Remove MCP server (never got working), add SearXNG Redis, update README/DEVLOG, clean start-stack.sh

Add in-container CUDA/GPU tuning: env vars for TabbyAPI EXL3 + Ollama, shared_memory 16g, Dockerfile ENV defaults, README tuning docs

Add Tabby_Tavern_Sell_Sheet.pdf

Clean deploy: Tabby-Tavern AI Stack - sanitized configs, no secrets, no model weights, no user data