[Misc] Remove #4789 workaround left in vllm/entrypoints/openai/run_batch.py (#5756)

vllm/entrypoints/openai/run_batch.py

 import asyncio
-import sys
 from io import StringIO
 from typing import Awaitable, List
 
@@ -137,9 +136,6 @@ async def main(args):
     output_buffer.seek(0)
     await write_file(args.output_file, output_buffer.read().strip())
 
-    # Temporary workaround for https://github.com/vllm-project/vllm/issues/4789
-    sys.exit(0)
-
 
 if __name__ == "__main__":
     args = parse_args()