test: add property-based tests and integration tests for retry-on-ratelimit

Add 302 property-based unit tests (proptest, 100+ iterations each) and
13 integration test scenarios covering all retry behaviors.

Unit tests cover:
- Configuration round-trip parsing, defaults, and validation
- Status code range expansion and error classification
- Exponential backoff formula, bounds, and scope filtering
- Provider selection strategy correctness and fallback ordering
- Retry-After state scope behavior and max expiration updates
- Cooldown exclusion invariants and initial selection cooldown
- Bounded retry (max_attempts + budget enforcement)
- Request preservation across retries
- Latency trigger sliding window and block state management
- Timeout vs high-latency precedence
- Error response detail completeness

Integration tests (tests/e2e/):
- IT-1 through IT-13 covering 429/503 retry, exhaustion, backoff,
  fallback priority, Retry-After honoring, timeout retry, high-latency
  failover, streaming preservation, and body preservation

tests/e2e/configs/retry_it10_timeout_triggers_retry.yaml

@@ -0,0 +1,27 @@
+version: v0.3.0
+
+listeners:
+  - type: model
+    name: model_listener
+    port: 12000
+
+model_providers:
+  - model: openai/gpt-4o
+    base_url: http://host.docker.internal:${MOCK_PRIMARY_PORT}
+    access_key: test-key-primary
+    default: true
+    retry_policy:
+      fallback_models: [anthropic/claude-3-5-sonnet]
+      default_strategy: "different_provider"
+      default_max_attempts: 2
+      on_status_codes:
+        - codes: [429]
+          strategy: "different_provider"
+          max_attempts: 2
+      on_timeout:
+        strategy: "different_provider"
+        max_attempts: 2
+
+  - model: anthropic/claude-3-5-sonnet
+    base_url: http://host.docker.internal:${MOCK_SECONDARY_PORT}
+    access_key: test-key-secondary

tests/e2e/configs/retry_it11_high_latency_failover.yaml

@@ -0,0 +1,33 @@
+version: v0.3.0
+
+listeners:
+  - type: model
+    name: model_listener
+    port: 12000
+
+model_providers:
+  - model: openai/gpt-4o
+    base_url: http://host.docker.internal:${MOCK_PRIMARY_PORT}
+    access_key: test-key-primary
+    default: true
+    retry_policy:
+      fallback_models: [anthropic/claude-3-5-sonnet]
+      default_strategy: "different_provider"
+      default_max_attempts: 2
+      on_status_codes:
+        - codes: [429]
+          strategy: "different_provider"
+          max_attempts: 2
+      on_high_latency:
+        threshold_ms: 1000
+        measure: "total"
+        min_triggers: 1
+        strategy: "different_provider"
+        max_attempts: 2
+        block_duration_seconds: 60
+        scope: "model"
+        apply_to: "global"
+
+  - model: anthropic/claude-3-5-sonnet
+    base_url: http://host.docker.internal:${MOCK_SECONDARY_PORT}
+    access_key: test-key-secondary

tests/e2e/configs/retry_it12_streaming.yaml

@@ -0,0 +1,23 @@
+version: v0.3.0
+
+listeners:
+  - type: model
+    name: model_listener
+    port: 12000
+
+model_providers:
+  - model: openai/gpt-4o
+    base_url: http://host.docker.internal:${MOCK_PRIMARY_PORT}
+    access_key: test-key-primary
+    default: true
+    retry_policy:
+      default_strategy: "different_provider"
+      default_max_attempts: 2
+      on_status_codes:
+        - codes: [429]
+          strategy: "different_provider"
+          max_attempts: 2
+
+  - model: anthropic/claude-3-5-sonnet
+    base_url: http://host.docker.internal:${MOCK_SECONDARY_PORT}
+    access_key: test-key-secondary

tests/e2e/configs/retry_it13_body_preserved.yaml

@@ -0,0 +1,23 @@
+version: v0.3.0
+
+listeners:
+  - type: model
+    name: model_listener
+    port: 12000
+
+model_providers:
+  - model: openai/gpt-4o
+    base_url: http://host.docker.internal:${MOCK_PRIMARY_PORT}
+    access_key: test-key-primary
+    default: true
+    retry_policy:
+      default_strategy: "different_provider"
+      default_max_attempts: 2
+      on_status_codes:
+        - codes: [429]
+          strategy: "different_provider"
+          max_attempts: 2
+
+  - model: anthropic/claude-3-5-sonnet
+    base_url: http://host.docker.internal:${MOCK_SECONDARY_PORT}
+    access_key: test-key-secondary

tests/e2e/configs/retry_it1_basic_429.yaml

@@ -0,0 +1,23 @@
+version: v0.3.0
+
+listeners:
+  - type: model
+    name: model_listener
+    port: 12000
+
+model_providers:
+  - model: openai/gpt-4o
+    base_url: http://host.docker.internal:${MOCK_PRIMARY_PORT}
+    access_key: test-key-primary
+    default: true
+    retry_policy:
+      default_strategy: "different_provider"
+      default_max_attempts: 2
+      on_status_codes:
+        - codes: [429]
+          strategy: "different_provider"
+          max_attempts: 2
+
+  - model: anthropic/claude-3-5-sonnet
+    base_url: http://host.docker.internal:${MOCK_SECONDARY_PORT}
+    access_key: test-key-secondary

tests/e2e/configs/retry_it2_503_different_provider.yaml

@@ -0,0 +1,23 @@
+version: v0.3.0
+
+listeners:
+  - type: model
+    name: model_listener
+    port: 12000
+
+model_providers:
+  - model: openai/gpt-4o
+    base_url: http://host.docker.internal:${MOCK_PRIMARY_PORT}
+    access_key: test-key-primary
+    default: true
+    retry_policy:
+      default_strategy: "different_provider"
+      default_max_attempts: 2
+      on_status_codes:
+        - codes: [503]
+          strategy: "different_provider"
+          max_attempts: 2
+
+  - model: anthropic/claude-3-5-sonnet
+    base_url: http://host.docker.internal:${MOCK_SECONDARY_PORT}
+    access_key: test-key-secondary

tests/e2e/configs/retry_it3_all_exhausted.yaml

@@ -0,0 +1,23 @@
+version: v0.3.0
+
+listeners:
+  - type: model
+    name: model_listener
+    port: 12000
+
+model_providers:
+  - model: openai/gpt-4o
+    base_url: http://host.docker.internal:${MOCK_PRIMARY_PORT}
+    access_key: test-key-primary
+    default: true
+    retry_policy:
+      default_strategy: "different_provider"
+      default_max_attempts: 2
+      on_status_codes:
+        - codes: [429]
+          strategy: "different_provider"
+          max_attempts: 2
+
+  - model: anthropic/claude-3-5-sonnet
+    base_url: http://host.docker.internal:${MOCK_SECONDARY_PORT}
+    access_key: test-key-secondary

tests/e2e/configs/retry_it4_no_retry_policy.yaml

@@ -0,0 +1,17 @@
+version: v0.3.0
+
+listeners:
+  - type: model
+    name: model_listener
+    port: 12000
+
+model_providers:
+  - model: openai/gpt-4o
+    base_url: http://host.docker.internal:${MOCK_PRIMARY_PORT}
+    access_key: test-key-primary
+    default: true
+    # No retry_policy — errors should be returned directly to client
+
+  - model: anthropic/claude-3-5-sonnet
+    base_url: http://host.docker.internal:${MOCK_SECONDARY_PORT}
+    access_key: test-key-secondary

tests/e2e/configs/retry_it5_max_attempts.yaml

@@ -0,0 +1,27 @@
+version: v0.3.0
+
+listeners:
+  - type: model
+    name: model_listener
+    port: 12000
+
+model_providers:
+  - model: openai/gpt-4o
+    base_url: http://host.docker.internal:${MOCK_PRIMARY_PORT}
+    access_key: test-key-primary
+    default: true
+    retry_policy:
+      default_strategy: "different_provider"
+      default_max_attempts: 1
+      on_status_codes:
+        - codes: [429]
+          strategy: "different_provider"
+          max_attempts: 1
+
+  - model: anthropic/claude-3-5-sonnet
+    base_url: http://host.docker.internal:${MOCK_SECONDARY_PORT}
+    access_key: test-key-secondary
+
+  - model: mistral/mistral-large
+    base_url: http://host.docker.internal:${MOCK_TERTIARY_PORT}
+    access_key: test-key-tertiary

tests/e2e/configs/retry_it6_backoff_delay.yaml

@@ -0,0 +1,24 @@
+version: v0.3.0
+
+listeners:
+  - type: model
+    name: model_listener
+    port: 12000
+
+model_providers:
+  - model: openai/gpt-4o
+    base_url: http://host.docker.internal:${MOCK_PRIMARY_PORT}
+    access_key: test-key-primary
+    default: true
+    retry_policy:
+      default_strategy: "same_model"
+      default_max_attempts: 3
+      on_status_codes:
+        - codes: [429]
+          strategy: "same_model"
+          max_attempts: 3
+      backoff:
+        apply_to: "same_model"
+        base_ms: 500
+        max_ms: 5000
+        jitter: false

tests/e2e/configs/retry_it7_fallback_priority.yaml

@@ -0,0 +1,28 @@
+version: v0.3.0
+
+listeners:
+  - type: model
+    name: model_listener
+    port: 12000
+
+model_providers:
+  - model: openai/gpt-4o
+    base_url: http://host.docker.internal:${MOCK_PRIMARY_PORT}
+    access_key: test-key-primary
+    default: true
+    retry_policy:
+      fallback_models: [anthropic/claude-3-5-sonnet, mistral/mistral-large]
+      default_strategy: "different_provider"
+      default_max_attempts: 3
+      on_status_codes:
+        - codes: [429]
+          strategy: "different_provider"
+          max_attempts: 3
+
+  - model: anthropic/claude-3-5-sonnet
+    base_url: http://host.docker.internal:${MOCK_FALLBACK1_PORT}
+    access_key: test-key-fallback1
+
+  - model: mistral/mistral-large
+    base_url: http://host.docker.internal:${MOCK_FALLBACK2_PORT}
+    access_key: test-key-fallback2

tests/e2e/configs/retry_it8_retry_after_honored.yaml

@@ -0,0 +1,23 @@
+version: v0.3.0
+
+listeners:
+  - type: model
+    name: model_listener
+    port: 12000
+
+model_providers:
+  - model: openai/gpt-4o
+    base_url: http://host.docker.internal:${MOCK_PRIMARY_PORT}
+    access_key: test-key-primary
+    default: true
+    retry_policy:
+      default_strategy: "same_model"
+      default_max_attempts: 2
+      on_status_codes:
+        - codes: [429]
+          strategy: "same_model"
+          max_attempts: 2
+      retry_after_handling:
+        scope: "model"
+        apply_to: "request"
+        max_retry_after_seconds: 300

tests/e2e/configs/retry_it9_retry_after_blocks_selection.yaml

@@ -0,0 +1,36 @@
+version: v0.3.0
+
+listeners:
+  - type: model
+    name: model_listener
+    port: 12000
+
+model_providers:
+  - model: openai/gpt-4o
+    base_url: http://host.docker.internal:${MOCK_PRIMARY_PORT}
+    access_key: test-key-primary
+    default: true
+    retry_policy:
+      fallback_models: [anthropic/claude-3-5-sonnet]
+      default_strategy: "different_provider"
+      default_max_attempts: 2
+      on_status_codes:
+        - codes: [429]
+          strategy: "different_provider"
+          max_attempts: 2
+      retry_after_handling:
+        scope: "model"
+        apply_to: "global"
+        max_retry_after_seconds: 300
+
+  - model: anthropic/claude-3-5-sonnet
+    base_url: http://host.docker.internal:${MOCK_SECONDARY_PORT}
+    access_key: test-key-secondary
+    default: false
+    retry_policy:
+      default_strategy: "different_provider"
+      default_max_attempts: 2
+      on_status_codes:
+        - codes: [429]
+          strategy: "different_provider"
+          max_attempts: 2