Remove unsupported ChatInterface parameters

Fix history format for ChatInterface

Switch to gr.ChatInterface for proper Gradio 6.5 compatibility

Fix Gradio Chatbot message format error

Optimize inference speed and fix Spaces compatibility

Add detailed progress tracking for model initialization

Load model once at startup instead of per query

Update bitsandbytes to >=0.46.1

Add 4-bit quantization and audioop-lts for Python 3.13