Enhance AssistantConfig and pipeline for FastGPT integration

- Add new fields in AssistantConfig for FastGPT connection details, including `fastgpt_api_url`, `fastgpt_api_key`, and `fastgpt_app_id`.
- Update the pipeline to utilize the new FastGPT configuration, ensuring proper integration with external services.
- Introduce type handling for different assistant types, including support for realtime modes and external brain management.
- Refactor frontend components to include hints for FastGPT configuration inputs, improving user guidance during setup.

backend/schemas.py

@@ -19,6 +19,11 @@ AssistantType = Literal["prompt", "workflow", "dify", "fastgpt", "opencode"]
 # 外部应用类型:其 config.apiKey 是该助手私有密钥,读时打码 / 写时哨兵
 EXTERNAL_TYPES = {"dify", "fastgpt", "opencode"}
 
+# 支持 realtime(语音到语音)的类型;外部托管大脑只能走 cascade。
+# 与 services.brains 各 BrainSpec.supported_runtime_modes 对齐(此处独立声明,
+# 避免 HTTP schema 层为做校验而引入 pipecat 重依赖)。
+REALTIME_CAPABLE_TYPES = {"prompt", "workflow"}
+
 
 class CamelModel(BaseModel):
     """JSON camelCase ↔ Python snake_case。protected_namespaces 关掉以允许 model_id。"""
@@ -67,6 +72,9 @@ class AssistantUpsert(CamelModel):
                 setattr(self, field, "")
         if "graph" not in allowed:
             self.graph = {}
+        # 外部托管大脑只能 cascade,拦住不兼容的 realtime
+        if self.runtime_mode == "realtime" and self.type not in REALTIME_CAPABLE_TYPES:
+            raise ValueError(f"类型 {self.type} 不支持 realtime 运行模式")
         return self