fix: Transition to cumulative yielding in Python backend and replacement rendering in JS frontend to prevent token duplication

fix: Add return type hint -> str to chat endpoint to prevent UserWarning and enable client data reception

fix: Rename custom config route to hf_token_status and align with server mode client connection docs

fix: Load highlight.js browser bundle and resolve template suggestion card string syntax error

fix: Optimize JS module loading, add readyState fallbacks, and make localStorage session loading robust

feat: Add gradio.Server backend and premium custom frontend

initial commit