Add brain memory services and APIs

Introduce the backend pieces for brain memory ingestion, routing, and
system telemetry so the new knowledge workflows can project data into a
brain view. The supporting tests lock in the new behavior and keep the
expanded backend surface stable.

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

backend/app/schemas/conversation.py

@@ -12,6 +12,7 @@ class MessageOut(BaseModel):
     content: str
     model: str | None
     tokens_used: int | None
+    attachments: list[dict] | None = None
     created_at: datetime
 
     model_config = {"from_attributes": True}
@@ -35,7 +36,8 @@ class ChatRequest(BaseModel):
     message: str
     conversation_id: str | None = None
     agent_id: str | None = None
-    file_ids: list[str] = []  # 新增
+    model_name: str | None = None
+    file_ids: list[str] = []
 
 
 class ChatResponse(BaseModel):
@@ -43,3 +45,4 @@ class ChatResponse(BaseModel):
     message_id: str
     content: str
     agent_name: str
+    model_name: str | None = None