update(gradio app): uplift the concurrency limit for the app

app.py

@@ -193,5 +193,5 @@ with gr.Blocks(fill_height=True) as demo:
     #             "\n(Note that it can be much slower than gpt-3.5, openai's api can be unstable sometimes.)\n"
     #             )
 
-demo.queue()
+demo.queue(default_concurrency_limit=None)
 demo.launch(share=False)