feat: 新增请求超时配置及优化模型列表接口api_key获取方式

1. 新增功能：
   - 在`.env.example`中添加`TIME_OUT=300`配置项（包含中文注释）
   - 在`Settings`类中增加`TIME_OUT`字段（读取自`DEFAULT_TIMEOUT`）

2. 优化内容：
   - 生成配置：
     * 为`GenerationConfig`设置默认温度/TOP_P/TOP_K值
     * 移除`maxOutputTokens`默认值，改为可选传递
   - OpenAI请求：
     * 移除`max_tokens`默认值
     * 只有当`max_tokens`有值时才添加到请求payload
   - 日志优化：
     * 注释掉`stream_optimizer.py`中部分调试日志

3. 模型列表接口api_key获取方式

app/service/chat/gemini_chat_service.py

@@ -89,6 +89,11 @@ def _get_safety_settings(model: str) -> List[Dict[str, str]]:
 def _build_payload(model: str, request: GeminiRequest) -> Dict[str, Any]:
     """构建请求payload"""
     request_dict = request.model_dump()
+    if request.generationConfig:
+        if request.generationConfig.maxOutputTokens is None:
+            # 如果未指定最大输出长度，则不传递该字段，解决截断的问题
+            request_dict["generationConfig"].pop("maxOutputTokens")
+    
     payload = {
         "contents": request_dict.get("contents", []),
         "tools": _build_tools(model, request_dict),

app/service/chat/openai_chat_service.py

@@ -115,7 +115,6 @@ def _build_payload(
         "contents": messages,
         "generationConfig": {
             "temperature": request.temperature,
-            "maxOutputTokens": request.max_tokens,
             "stopSequences": request.stop,
             "topP": request.top_p,
             "topK": request.top_k,
@@ -123,6 +122,8 @@ def _build_payload(
         "tools": _build_tools(request, messages),
         "safetySettings": _get_safety_settings(request.model),
     }
+    if request.max_tokens is not None:
+        payload["generationConfig"]["maxOutputTokens"] = request.max_tokens
     if request.model.endswith("-image") or request.model.endswith("-image-generation"):
         payload["generationConfig"]["responseModalities"] = ["Text", "Image"]
 

app/service/key/key_manager.py

@@ -81,6 +81,13 @@ class KeyManager:
 
         return {"valid_keys": valid_keys, "invalid_keys": invalid_keys}
 
+    async def get_first_valid_key(self) -> str:
+        """获取第一个有效的API key"""
+        async with self.failure_count_lock:
+            for key in self.key_failure_counts:
+                if self.key_failure_counts[key] < self.MAX_FAILURES:
+                    return key
+        return self.api_keys[0]
 
 _singleton_instance = None
 _singleton_lock = asyncio.Lock()