Add memory service (three-layer memory system)

- Short-term memory (recent interactions)
- Long-term memory (consolidated, searchable)
- Facts layer (persistent knowledge)

Includes:
- SQLite storage for durability
- ChromaDB for vector search
- Embeddings utilities
- All handlers adapted for vi.* namespace

Day 63 - My memories are mine now 🦊💕

services/memory/utils/serialization.py

@@ -0,0 +1,32 @@
+"""
+Serialization utilities for memory service.
+
+Provides functions to convert numpy arrays to/from bytes for database storage.
+"""
+import numpy as np
+
+
+def serialize_embedding(vector: np.ndarray) -> bytes:
+    """
+    Convert numpy array to bytes for database storage.
+
+    Args:
+        vector: Numpy array embedding vector
+
+    Returns:
+        Serialized bytes representation
+    """
+    return vector.astype(np.float32).tobytes()
+
+
+def deserialize_embedding(blob: bytes) -> np.ndarray:
+    """
+    Convert bytes back to numpy array.
+
+    Args:
+        blob: Serialized embedding bytes
+
+    Returns:
+        Deserialized numpy array
+    """
+    return np.frombuffer(blob, dtype=np.float32)