PDF Extraction (pdf_extractor.py) — Uses PyMuPDF to extract text spans (with position, font, and style metadata), images, and tables. Classifies each page as digital (has selectable text) or scanned ...