feat: implement three-tier metabolic LLM router (Refs #966)

- Add MetabolicRouter routing tasks to T1/T2/T3 model tiers based on
  complexity classification
- classify_complexity() assigns cheapest-sufficient tier using keyword
  matching and game state signals (stuck, active quests, dialogue)
- build_prompt() assembles OpenAI-compatible message lists from game/UI state
- T3 (Qwen3-32B) pauses the game world before inference and unpauses in
  a finally block (graceful if no world adapter)
- Fix: narrow "what should" phrase to "what should i do" to avoid
  misclassifying dialogue tasks as T3
- Add pytestmark=pytest.mark.unit so tests run in tox -e unit
- Export all public symbols via infrastructure/router/__init__.py

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

src/infrastructure/router/metabolic.py

@@ -62,7 +62,7 @@ _T3_KEYWORDS = frozenset(
         "recover", "multi-step", "long-term", "negotiate", "persuade",
         "faction", "reputation", "best", "optimal", "recommend",
         "analyze", "analyse", "evaluate", "decide", "complex", "how do i",
-        "what should", "help me figure", "what is the best",
+        "what should i do", "help me figure", "what is the best",
     }
 )
 

tests/infrastructure/test_metabolic_router.py

@@ -5,6 +5,8 @@ from unittest.mock import AsyncMock, MagicMock, patch
 
 import pytest
 
+pytestmark = pytest.mark.unit
+
 from infrastructure.router.metabolic import (
     DEFAULT_TIER_MODELS,
     MetabolicRouter,