- PPTX: Add Table/Group/Shape extraction for complete content
- DOCX: Extract Base64 images to RawContent.Images with placeholders
- PDF: Improve table parsing with gap-based column detection
- PDF: Add page number filtering for cleaner text output
- Parser: Fix metadata key mapping (Title/Author across formats)