Fix VectorDB search: use content field instead of text

The vector-db API returns message content in a top-level "content" field
with "author" and "channel" also top-level (not nested under metadata).
Previous code read "text" and "metadata.author" which returned empty strings,
making all vector search results invisible to the LLM.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

wiki-vector-chat.py

@@ -152,12 +152,11 @@ async def search_vector(query: str, top_k: int = 5) -> str:
                 return ""
             lines = []
             for h in hits[:top_k]:
-                text = h.get("text", "")[:300]
+                text = h.get("content", "") or h.get("text", "")
                 score = h.get("score", 0)
                 source = h.get("source", "unknown")
-                meta = h.get("metadata", {})
-                author = meta.get("author", "")
-                channel = meta.get("channel", "")
+                author = h.get("author", "")
+                channel = h.get("channel", "")
                 preview = text.replace("\n", " ")[:200]
                 lines.append(f"[{source}] @{author} in #{channel}: {preview}  (score: {score:.2f})")
             return "\n\n".join(lines)