perf: fully async DB - results in 5-7s, background check+save

perf: async DB save with duplicate check for faster extraction

feat: PDF preview, database integration, and improved error handling

refactor: remove obsolete OCR test file and enhance address extraction logic

Refactor: Replace Tesseract with DocTR and integrate LayoutLMv3-DocTR model

feat: add AI detection overlay visualization with bounding boxes on extracted entities

chore: Add scaffolded database layer (not yet implemented)

Restore full history with LFS images and all fixes

feat: Update Dockerfile and requirements for PDF processing, add new dependencies, and refactor API structure

feat: Enhance pipeline with smart PDF handling, Pydantic validation, and semantic hashing, and refactor API to src.

refactor: Reorganize project structure

feat: Add .dockerignore, enhance UI to display receipt number and robustly handle bill-to, and update README with an additional dataset.

Updated Extraction Logic for more Robustness

feat: Implement robust OCR, and cross-platform support

feat: Add Phase 3 generalization scripts and clean up legacy files

feat: LayoutLMv3 integration, Streamlit UI toggle, README refresh, .gitignore

Complete Version 0.5 with Streamlit UI and full pipeline