feat(llm): add Gemini CLI provider as primary; set qwen3.5:9b as default Ollama model

- Add GeminiCliProvider: shells out to `gemini -p` with --output-format json,
  injection-safe prompt passing, MCP server suppression via temp workdir,
  6-slot concurrency semaphore, 60s subprocess deadline
- Add --llm-provider, --llm-model, --llm-base-url CLI flags for per-call overrides
- Provider chain: Gemini CLI → OpenAI → Ollama → Anthropic
- Move LLM timing to dispatch layer (LLM: Xs on stderr)
- Default Ollama model: qwen3:8b → qwen3.5:9b (benchmark shows better schema extraction)
- Add noxa mcp subcommand
- Add docs/reports/llm-benchmark-2026-04-11.md (Gemini vs qwen3.5:4b vs qwen3.5:9b)
- Bump version 0.3.11 → 0.4.0

Co-authored-by: Claude <claude@anthropic.com>

Cargo.lock

@@ -1746,7 +1746,7 @@ dependencies = [
 
 [[package]]
 name = "noxa-cli"
-version = "0.3.11"
+version = "0.4.0"
 dependencies = [
  "clap",
  "dotenvy",
@@ -1768,7 +1768,7 @@ dependencies = [
 
 [[package]]
 name = "noxa-core"
-version = "0.3.11"
+version = "0.4.0"
 dependencies = [
  "ego-tree",
  "once_cell",
@@ -1786,7 +1786,7 @@ dependencies = [
 
 [[package]]
 name = "noxa-fetch"
-version = "0.3.11"
+version = "0.4.0"
 dependencies = [
  "bytes",
  "calamine",
@@ -1808,7 +1808,7 @@ dependencies = [
 
 [[package]]
 name = "noxa-llm"
-version = "0.3.11"
+version = "0.4.0"
 dependencies = [
  "async-trait",
  "jsonschema",
@@ -1822,7 +1822,7 @@ dependencies = [
 
 [[package]]
 name = "noxa-mcp"
-version = "0.3.11"
+version = "0.4.0"
 dependencies = [
  "dirs",
  "dotenvy",
@@ -1843,7 +1843,7 @@ dependencies = [
 
 [[package]]
 name = "noxa-pdf"
-version = "0.3.11"
+version = "0.4.0"
 dependencies = [
  "pdf-extract",
  "thiserror",