revert: Restore original Gradio UI from before conversation

feat: Full-screen chatbot interface with example questions

feat: Enable conversational responses for ANY question

feat: Streamline Gradio UI with 4 major improvements

fix: Add debug logging and follow-up question support

feat: Speed & intelligence overhaul - 3x faster with natural responses

fix: Handle ChatCompletionMessage objects in message trimming

fix: Strengthen system prompt to eliminate tool call announcements

fix: Improve UX - text comparisons, remove preset bloat, fix tool message errors

fix: Extract nested funding data in grant comparisons

fix: Add aggressive timeout prevention and context limits

perf: Optimize latency and reduce token usage across QA and search services

fix: Update import path in app.py for HF deployment

feat: Major scraper enhancements - consistent format, better extraction, per-month costs

fix: Remove temperature parameter for GPT-5 compatibility (only supports default)

fix: Update .dockerignore to include data/snapshots for HF deployment

fix: Change all max_tokens to max_completion_tokens for GPT-5 compatibility

debug: Add comprehensive directory listing to diagnose missing data on HF

debug: Add comprehensive path diagnostics for HF deployment

debug: Add more detailed logging for grant initialization

debug: Add logging to diagnose grant loading issue on HF

fix: Change max_tokens to max_completion_tokens for GPT-5 API compatibility

feat: Merge GPT-5 enhancements with working c72a240 base + restore crawler

fix: Complete restoration of analyzer/ directory from c72a240

fix: Restore exact Gradio UI from working commit c72a240

fix: Restore exact working structure from commit c72a240

fix: Restore working Hugging Face deployment structure

chore: Remove websocket dependencies - not used in current version

fix: Restore working app.py structure from pre-merge version

fix: Restore simple relative path for grant loading

feat: Add comprehensive logging to app.py startup

fix: Remove WebSocket/SSE streaming UI and add detailed logging

fix: Use absolute path for loading grant snapshots

fix: Remove all remaining emojis from demo_app UI

fix: Add websocket-client and gradio to requirements

chore: Remove old grant-radar-hf-deploy directory and duplicate snapshots

fix: Add app.py wrapper for Hugging Face Spaces compatibility

fix: Update app_file to point to demo_app.py

fix: Add Hugging Face YAML metadata to README.md

chore: Add venv to gitignore

chore: Remove large files from git tracking

chore: Configure Git LFS for large files

feat: Major system enhancements - GPT-5 support, monitoring, translation, and optimizations

Added Dev Container Folder

fix: uncomment data/ in gitignore so grant data is deployed to Streamlit Cloud

debug: add logging to track grant loading in initialize_app

fix: clarify system prompt - 'list all grants' should use list_grants, not get_all_grant_summaries

fix: use stronger system prompt from demo_app to force tool execution

fix: make hybrid search index optional in chat_tools - graceful fallback if scikit-learn unavailable

fix: make hybrid search index import optional - graceful fallback if scikit-learn unavailable