Extract text, blocks, tables, and page data from a PDF page or image. Use it for OCR, layout-aware parsing, and table capture.