feat: switch GLM-5.2 provider to fireworks-ai

feat: standard chat mode should not have terminal-like look

chore: remove unused sidebar buttons and restrict workspace to Standard and Terminal modes

chore: remove hf token from UI configuration and rely on backend environment variable

feat: use AsyncOpenAI and set X-Accel-Buffering headers for instant streaming

feat: call the app GLM 5.2 and add terminal dashboard & status bar elements

chore: remove other models and add requirements.txt

feat: Add GLM-5.2 terminal-style chat UI using gradio.Server

initial commit