Extract PDF content

Extracts text, tables, and images from a PDF using AI-powered content recognition. Returns structured JSON with content elements preserving reading order and layout relationships.