feat(02-01): add two-layer memory system — Redis sliding window + pgvector long-term

- ConversationEmbedding ORM model with Vector(384) column (pgvector)
- memory_short_key, escalation_status_key, pending_tool_confirm_key in redis_keys.py
- orchestrator/memory/short_term.py: RPUSH/LTRIM sliding window (get_recent_messages, append_message)
- orchestrator/memory/long_term.py: pgvector HNSW cosine search (retrieve_relevant, store_embedding)
- Migration 002: conversation_embeddings table, HNSW index, RLS with FORCE, SELECT/INSERT only
- 10 unit tests (fakeredis), 6 integration tests (pgvector) — all passing
- Auto-fix [Rule 3]: postgres image updated to pgvector/pgvector:pg16 (extension required)

packages/shared/pyproject.toml

@@ -19,6 +19,7 @@ dependencies = [
     "httpx>=0.28.0",
     "slowapi>=0.1.9",
     "bcrypt>=4.0.0",
+    "pgvector>=0.3.0",
 ]
 
 [tool.hatch.build.targets.wheel]