Tighten should_knock_early() so AI no longer knocks with projected
scores of 12-14. New range: max_acceptable 5-9 (was 8-18), with
scaled knock_chance by score quality and an exception when all
opponents show 25+ visible points.
Fix 5 pre-existing test failures:
- test_event_replay: use game.current_player() instead of hardcoding
"p1", since dealer logic makes p2 go first
- game.py: include current_player_idx in round_started event so state
replay knows the correct starting player
- test_house_rules: rename test_rule_config → run_rule_config so
pytest doesn't collect it as a test fixture
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Phase 1 - Critical Fixes:
- Add game_lock (asyncio.Lock) to Room class for serializing mutations
- Wrap all game action handlers in lock to prevent race conditions
- Split Card.to_dict into to_dict (full data) and to_client_dict (hidden)
- Fix CardState.from_dict to handle missing rank/suit gracefully
- Fix GameOptions reconstruction in recovery_service (dict -> object)
- Extend state cache TTL from 4h to 24h, add touch_game method
Phase 2 - Security:
- Add optional WebSocket authentication via token query param
- Use authenticated user ID/name when available
- Add auth support to spectator WebSocket endpoint
Phase 3 - Performance:
- Make stats processing async (fire-and-forget) to avoid blocking
game completion notifications
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
