v43: Improve VIN extraction with Korean transliteration and bbox zone detection

v41: Fix owner name extraction for spaced OCR text

v40: Add purchase price (출고가격) field extraction

v39: Add model-spec lookup for dimension fill/correction

v38: Revert to single Korean OCR engine - fix cpu-basic timeout

v37: Fix race condition - thread-safe OCR engine initialization

v35: Skip English engine warmup to prevent startup timeout on cpu-basic

v34: Dual OCR engine (Korean + English) for character-type-aware recognition

v33: Improve OCR accuracy - bbox extraction, Korean fuel types, VIN year, date

v32: Disable OCR in PP-Structure to reduce startup time

v30: Fix connection timeout - warmup engines at startup, restore ssr=False

v29: Fix PP-Structure lang error - use 'ch' instead of 'korean' for layout

v28: Add PP-Structure layout analysis engine for form-aware OCR

v24: Port OCR_Vehicle_02 algorithms - preprocessing, bbox extraction, standards

v23: Fix fuel detection - Hydrogen before LPG, add H2/FCEV keywords

v22: Fix owner name cleanup for bracketless OCR variants

v21: Fix owner name - use next-line extraction for table layout

v20: Fix owner name extraction using ⑨ marker and robust label matching

v19: Fix owner name cleanup, model name extraction, weight/capacity parsing

v18: ISO 3779 VIN validation, OCR correction, and owner name cleanup

feat: improve VIN extraction and show OCR debug in summary (v17)

feat: major parser rewrite for robust field extraction (v16)

fix: broaden document type verification and show OCR preview on skip (v15)

fix: add debug diagnostics for PaddleOCR 3.x result format (v13)

fix: robust PaddleOCR 3.x result parsing with multi-strategy fallback (v11)

fix: show detailed error messages in UI for debugging

fix: auto-detect PaddleOCR version and use correct API

Remove dead code and fix issues from code review

feat: add core processing pipeline (processor.py) and tests

feat: add parse_single() method to CarRegistrationParser

refactor: simplify config for HF Spaces deployment

chore: remove unused files and __pycache__ from tracking

Downgrade PaddleOCR from 3.3+ to 2.x API

Add structured parser, template extractor, and fix OCR processing issues

Remove .DS_Store files and update .gitignore

Migrate to local OCR (PaddleOCR) and add speed optimizations

Initial commit