1.2: Fix thread safety in memory.py embedding DB

Replace 4 standalone sqlite3.connect()/conn.close() pairs with a
thread-local _embed_conn property, matching the pattern in db.py.
Adds WAL mode for better concurrent read/write performance.

This prevents potential collisions between scheduler threads and
Gradio request threads accessing the embedding database.

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

cheddahbot/memory.py

@@ -30,6 +30,7 @@ class MemorySystem:
         self._embedder = None
         self._embed_lock = threading.Lock()
         self._embed_db_path = self.memory_dir / "embeddings.db"
+        self._embed_local = threading.local()
         self._init_embed_db()
 
     # ── Public API ──
@@ -163,17 +164,23 @@ class MemorySystem:
 
     # ── Private: Embedding system ──
 
+    @property
+    def _embed_conn(self) -> sqlite3.Connection:
+        """Thread-local SQLite connection for embeddings DB (matches db.py pattern)."""
+        if not hasattr(self._embed_local, "conn"):
+            self._embed_local.conn = sqlite3.connect(str(self._embed_db_path))
+            self._embed_local.conn.execute("PRAGMA journal_mode=WAL")
+        return self._embed_local.conn
+
     def _init_embed_db(self):
-        conn = sqlite3.connect(str(self._embed_db_path))
-        conn.execute("""
+        self._embed_conn.execute("""
             CREATE TABLE IF NOT EXISTS embeddings (
                 id      TEXT PRIMARY KEY,
                 text    TEXT NOT NULL,
                 vector  BLOB NOT NULL
             )
         """)
-        conn.commit()
-        conn.close()
+        self._embed_conn.commit()
 
     def _get_embedder(self):
         if self._embedder is not None:
@@ -200,18 +207,14 @@ class MemorySystem:
         if embedder is None:
             return
         vec = embedder.encode([text])[0]
-        conn = sqlite3.connect(str(self._embed_db_path))
-        conn.execute(
+        self._embed_conn.execute(
             "INSERT OR REPLACE INTO embeddings (id, text, vector) VALUES (?, ?, ?)",
             (doc_id, text, vec.tobytes()),
         )
-        conn.commit()
-        conn.close()
+        self._embed_conn.commit()
 
     def _vector_search(self, query_vec: np.ndarray, top_k: int) -> list[dict]:
-        conn = sqlite3.connect(str(self._embed_db_path))
-        rows = conn.execute("SELECT id, text, vector FROM embeddings").fetchall()
-        conn.close()
+        rows = self._embed_conn.execute("SELECT id, text, vector FROM embeddings").fetchall()
 
         if not rows:
             return []
@@ -228,10 +231,8 @@ class MemorySystem:
         return scored[:top_k]
 
     def _clear_embeddings(self):
-        conn = sqlite3.connect(str(self._embed_db_path))
-        conn.execute("DELETE FROM embeddings")
-        conn.commit()
-        conn.close()
+        self._embed_conn.execute("DELETE FROM embeddings")
+        self._embed_conn.commit()
 
     def _fallback_search(self, query: str, top_k: int) -> list[dict]:
         """Simple keyword search when embeddings are unavailable."""