use plano-orchestrator for LLM routing, remove arch-router

Replace RouterService/RouterModelV1 (arch-router prompt) with
OrchestratorService/OrchestratorModelV1 (plano-orchestrator prompt)
for LLM routing. This ensures the correct system prompt is used when
llm_routing_model points at a Plano-Orchestrator model.

- Extend OrchestratorService with session caching, ModelMetricsService,
  top-level routing preferences, and determine_route() for LLM routing
- Delete RouterService, RouterModel trait, RouterModelV1, and
  ARCH_ROUTER_V1_SYSTEM_PROMPT
- Unify defaults to Plano-Orchestrator / plano-orchestrator
- Update CLI config generator, demos, docs, and config schema

Made-with: Cursor

demos/llm_routing/model_routing_service/vllm-deployment.yaml

@@ -1,18 +1,18 @@
 apiVersion: apps/v1
 kind: Deployment
 metadata:
-  name: arch-router
+  name: plano-orchestrator
   labels:
-    app: arch-router
+    app: plano-orchestrator
 spec:
   replicas: 1
   selector:
     matchLabels:
-      app: arch-router
+      app: plano-orchestrator
   template:
     metadata:
       labels:
-        app: arch-router
+        app: plano-orchestrator
     spec:
       tolerations:
         - key: nvidia.com/gpu
@@ -53,7 +53,7 @@ spec:
             - "--tokenizer"
             - "katanemo/Arch-Router-1.5B"
             - "--served-model-name"
-            - "Arch-Router"
+            - "Plano-Orchestrator"
             - "--gpu-memory-utilization"
             - "0.3"
             - "--tensor-parallel-size"
@@ -94,10 +94,10 @@ spec:
 apiVersion: v1
 kind: Service
 metadata:
-  name: arch-router
+  name: plano-orchestrator
 spec:
   selector:
-    app: arch-router
+    app: plano-orchestrator
   ports:
     - name: http
       port: 10000