Fix pgvector embedding format in recommender query

profile.embedding was being passed as str(numpy_array) which produces
scientific notation format. pgvector needs [n1,n2,...] format. Now
explicitly formats as comma-separated float list.

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

src/haunt_fm/services/recommender.py

@@ -53,9 +53,13 @@ async def get_recommendations(
             LIMIT :limit
         """)
 
+    # Format embedding as pgvector literal: [n1,n2,...]
+    emb = profile.embedding
+    emb_str = "[" + ",".join(str(float(x)) for x in emb) + "]"
+
     result = await session.execute(
         query,
-        {"profile_embedding": str(profile.embedding), "limit": limit},
+        {"profile_embedding": emb_str, "limit": limit},
     )
 
     return [