[Core] Shut down aDAG workers with clean async llm engine exit (#7224)

Signed-off-by: Rui Qiao <ruisearch42@gmail.com>

vllm/entrypoints/openai/rpc/server.py

@@ -36,6 +36,7 @@ class AsyncEngineRPCServer:
         """Cleanup all resources."""
         self.socket.close()
         self.context.destroy()
+        self.engine.shutdown_background_loop()
 
     async def get_model_config(self, identity):
         """Send the ModelConfig"""