Refactor: move mcp connection utilities to common (#11304)

### What problem does this PR solve?

As title

### Type of change

- [x] Refactoring

---------

Signed-off-by: Jin Hai <haijin.chn@gmail.com>

rag/llm/chat_model.py

@@ -22,7 +22,6 @@ import re
 import time
 from abc import ABC
 from copy import deepcopy
-from typing import Any, Protocol
 from urllib.parse import urljoin
 
 import json_repair
@@ -65,10 +64,6 @@ LENGTH_NOTIFICATION_CN = "······\n由于大模型的上下文窗口大小
 LENGTH_NOTIFICATION_EN = "...\nThe answer is truncated by your chosen LLM due to its limitation on context length."
 
 
-class ToolCallSession(Protocol):
-    def tool_call(self, name: str, arguments: dict[str, Any]) -> str: ...
-
-
 class Base(ABC):
     def __init__(self, key, model_name, base_url, **kwargs):
         timeout = int(os.environ.get("LM_TIMEOUT_SECONDS", 600))