Multi-Modal Chat

Text + image inputs, vision-aware responses.

• → Vision LLM integration with image-aware context
• → React + FastAPI full-stack with streaming responses
• → Image preprocessing and base64 handling