feat(system):优化后端

1.新增后端测试
2.增加了后端的加密
3.增加了i18n（国际化）

kilostar/api/chat.py

@@ -16,10 +16,40 @@ from fastapi import APIRouter, Depends
 from pydantic import BaseModel
 from kilostar.utils.ray_hook import ray_actor_hook
 from kilostar.utils.access import Accessor, TokenData
+from kilostar.core.individual.regulatory_node.template import (
+    MessageRequest,
+    MessageResponse,
+)
 
 chat_router = APIRouter(prefix="/api/v1/chat", tags=["chat"])
 
 
+def _extract_reply(resp: MessageResponse | None) -> str | None:
+    """从 RegulatoryNode.working 的输出里取出对用户的回复文本。
+
+    RegulatoryNode 现在的 output_type 只剩 ``MessageResponse``（聊天/简单任务/汇报），
+    没有则视为节点降级为静默——上层不写回 chat history。
+    """
+    if resp is None:
+        return None
+    return resp.reply_message
+
+
+async def _ask_regulatory(
+    *, user_id: str, chat_id: str, message: str
+) -> str | None:
+    """统一封装 chat 入口对 RegulatoryNode 的调用。"""
+    regulatory_node = ray_actor_hook("regulatory_node").regulatory_node
+    payload = MessageRequest(
+        platform="client",
+        user_name=user_id,
+        platform_id=chat_id,
+        message=message,
+    )
+    resp: MessageResponse | None = await regulatory_node.working.remote(payload)
+    return _extract_reply(resp)
+
+
 class CreateChatRequest(BaseModel):
     title: str = "新对话"
     initial_message: str
@@ -45,9 +75,7 @@ async def create_chat_session(
     )
 
     # 调用监管节点处理简单任务/交流
-    regulatory_node = ray_actor_hook("regulatory_node").regulatory_node
-    # 在此发起任务并等待或异步返回结果
-    response_msg = await regulatory_node.handle_chat_message.remote(
+    response_msg = await _ask_regulatory(
         user_id=token_data.user_id,
         chat_id=chat.chat_id,
         message=request.initial_message,
@@ -95,9 +123,10 @@ async def send_chat_message(
     )
 
     # 调用监管节点
-    regulatory_node = ray_actor_hook("regulatory_node").regulatory_node
-    response_msg = await regulatory_node.handle_chat_message.remote(
-        user_id=token_data.user_id, chat_id=chat_id, message=request.message
+    response_msg = await _ask_regulatory(
+        user_id=token_data.user_id,
+        chat_id=chat_id,
+        message=request.message,
     )
 
     # 存回复