clean api code, remove redundant background task. (#1102)

vllm/entrypoints/api_server.py

@@ -2,7 +2,7 @@ import argparse
 import json
 from typing import AsyncGenerator
 
-from fastapi import BackgroundTasks, FastAPI, Request
+from fastapi import FastAPI, Request
 from fastapi.responses import JSONResponse, Response, StreamingResponse
 import uvicorn
 
@@ -44,14 +44,8 @@ async def generate(request: Request) -> Response:
             ret = {"text": text_outputs}
             yield (json.dumps(ret) + "\0").encode("utf-8")
 
-    async def abort_request() -> None:
-        await engine.abort(request_id)
-
     if stream:
-        background_tasks = BackgroundTasks()
-        # Abort the request if the client disconnects.
-        background_tasks.add_task(abort_request)
-        return StreamingResponse(stream_results(), background=background_tasks)
+        return StreamingResponse(stream_results())
 
     # Non-streaming case
     final_output = None