feat: upgrade to qwen3.5, self-hosted Gitea CI, optimize Docker image

Model upgrade:
- qwen2.5:14b → qwen3.5:latest across config, tools, and docs
- Added qwen3.5 to multimodal model registry

Self-hosted Gitea CI:
- .gitea/workflows/tests.yml: lint + test jobs via act_runner
- Unified Dockerfile: pre-baked deps from poetry.lock for fast CI
- sitepackages=true in tox for ~2s dep resolution (was ~40s)
- OLLAMA_URL set to dead port in CI to prevent real LLM calls

Test isolation fixes:
- Smoke test fixture mocks create_timmy (was hitting real Ollama)
- WebSocket sends initial_state before joining broadcast pool (race fix)
- Tests use settings.ollama_model/url instead of hardcoded values
- skip_ci marker for Ollama-dependent tests, excluded in CI tox envs

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

src/infrastructure/models/multimodal.py

@@ -93,6 +93,18 @@ KNOWN_MODEL_CAPABILITIES: dict[str, set[ModelCapability]] = {
         ModelCapability.VISION,
     },
     # Qwen series
+    "qwen3.5": {
+        ModelCapability.TEXT,
+        ModelCapability.TOOLS,
+        ModelCapability.JSON,
+        ModelCapability.STREAMING,
+    },
+    "qwen3.5:latest": {
+        ModelCapability.TEXT,
+        ModelCapability.TOOLS,
+        ModelCapability.JSON,
+        ModelCapability.STREAMING,
+    },
     "qwen2.5": {
         ModelCapability.TEXT,
         ModelCapability.TOOLS,
@@ -259,6 +271,7 @@ DEFAULT_FALLBACK_CHAINS: dict[ModelCapability, list[str]] = {
     ],
     ModelCapability.TOOLS: [
         "llama3.1:8b-instruct",  # Best tool use
+        "qwen3.5:latest",  # Qwen 3.5 — strong tool use
         "llama3.2:3b",  # Smaller but capable
         "qwen2.5:7b",  # Reliable fallback
     ],