Implement Phase 4~14: LangGraph Agent, RAG pipeline, Gradio Web UI, voice interface

- Upgrade LLM to Qwen3-14B-4bit with Thinking mode (MlxChatModel as LangChain BaseChatModel)
- Add LangGraph ReAct agent with tool calling loop (search_documents, web_search, get_current_date, remember/recall_user_info)
- Add RAG pipeline: BAAI/bge-m3 embeddings + Qdrant vector store + semantic chunking (SemanticSplitter via cosine similarity)
- Replace fixed-size RecursiveCharacterTextSplitter with meaning-based SemanticSplitter (numpy only, no extra deps)
- Add Gradio Web UI (app.py): chat, document ingestion, document management tabs
- Add multi-user support (user_id isolation in DB + per-user agent cache + dropdown selector)
- Add conversation history restore from MySQL on agent init (Phase 11)
- Add UserProfileRepository for persistent user profile (remember/recall tools)
- Add thread-local DB connections to fix pymysql thread-safety with LangGraph ToolNode
- Add Phase 14 voice interface: Whisper STT (microphone → text) + macOS TTS (say -v Yuna)
- Enforce search_documents-first policy in system prompt and tool descriptions
- Update ROADMAP2.md: Phase 14 완료, Phase 13 청킹 부분 완료

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

services/rag/retriever_service.py

@@ -0,0 +1,67 @@
+from langchain_core.documents import Document
+from langchain_qdrant import QdrantVectorStore
+from qdrant_client import QdrantClient
+from qdrant_client.models import Filter, FieldCondition, MatchValue, FilterSelector
+
+
+class RetrieverService:
+    """Qdrant 벡터 검색 서비스. LangGraph Tool 및 직접 검색 모두 지원."""
+
+    def __init__(
+        self,
+        embeddings,
+        qdrant_url: str,
+        collection_name: str,
+        top_k: int,
+    ):
+        self._client = QdrantClient(url=qdrant_url)
+        self._collection_name = collection_name
+        self._store = QdrantVectorStore(
+            client=self._client,
+            collection_name=collection_name,
+            embedding=embeddings,
+        )
+        self._top_k = top_k
+
+    def as_retriever(self):
+        return self._store.as_retriever(search_kwargs={"k": self._top_k})
+
+    def search(self, query: str) -> list[Document]:
+        return self._store.similarity_search(query, k=self._top_k)
+
+    def list_documents(self) -> list[str]:
+        """Qdrant에 저장된 고유 파일 경로 목록을 반환한다."""
+        sources: set[str] = set()
+        offset = None
+        while True:
+            results, next_offset = self._client.scroll(
+                collection_name=self._collection_name,
+                with_payload=True,
+                limit=200,
+                offset=offset,
+            )
+            for point in results:
+                src = (point.payload or {}).get("metadata", {}).get("source", "")
+                if src:
+                    sources.add(src)
+            if next_offset is None:
+                break
+            offset = next_offset
+        return sorted(sources)
+
+    def delete_document(self, source: str) -> None:
+        """파일 경로로 저장된 모든 청크를 Qdrant에서 삭제한다."""
+        try:
+            self._client.delete(
+                collection_name=self._collection_name,
+                points_selector=FilterSelector(
+                    filter=Filter(
+                        must=[FieldCondition(
+                            key="metadata.source",
+                            match=MatchValue(value=source),
+                        )]
+                    )
+                ),
+            )
+        except Exception:
+            pass