fix: 后端 API 修复

- deps.py: 依赖注入优化
- matches.py: 比赛列表查询修复
- predict.py: 预测接口参数校验修复
- orchestrator.py: 多 Agent 编排逻辑修复

src/api/routes/predict.py

@@ -8,11 +8,11 @@ from __future__ import annotations
 
 import logging
 
-from fastapi import APIRouter, Depends, HTTPException, Query
+from fastapi import APIRouter, Depends, HTTPException, Query, Request
 from sqlalchemy import select
 from sqlalchemy.orm import selectinload
 
-from src.api.deps import rate_limit_predict, require_admin
+from src.api.deps import get_predict_rate_limit_remaining, rate_limit_predict, require_admin
 from src.api.schemas import PredictOut, PredictRequest, PredictionOut
 from src.db.base import AsyncSession, get_db_read, short_read
 from src.db.models import Match, Prediction
@@ -24,8 +24,8 @@ router = APIRouter(prefix="/api/v1", tags=["predict"])
 
 
 @router.post("/predict", response_model=PredictOut, dependencies=[Depends(rate_limit_predict)])
-async def predict(req: PredictRequest):
-    """对一场比赛调 LLM 预测。mode=multi(默认,5专家+终裁)或 single。
+async def predict(req: PredictRequest, request: Request):
+    """对一场比赛调 LLM 预测。mode=multi(默认,5专家+终裁)、single 或 baseline。
 
     公开接口,仅做限流保护(不要求登录)。
 
@@ -100,7 +100,7 @@ async def predict(req: PredictRequest):
         latency_ms=result.get("latency_ms", 0) if result_dict else result.latency_ms,
         prompt_tokens=result.get("prompt_tokens") if result_dict else getattr(result, "prompt_tokens", None),
         completion_tokens=result.get("completion_tokens") if result_dict else getattr(result, "completion_tokens", None),
-        rate_limit_remaining=_predict_limiter.remaining(get_client_ip(request)),
+        rate_limit_remaining=get_predict_rate_limit_remaining(request),
     )