🐜 Correct PDF creator metadata and support Dublin Core tags

[JFWooten4]

Summary

• Keep the human author in PDF /Author and XMP dc:creator, without incorrectly writing that name into /Creator or xmp:CreatorTool.
• Preserve the original authoring application when identifiable; omit an unknown creator rather than fabricating one. Continue identifying PDF Magic Enhancer as the producer.
• Synchronize descriptive metadata between PDF Info and Dublin Core/XMP: title, descriptions, keywords/tags, publisher, rights, language, and identifier. Retain source metadata across OCR and compression where supplied.
• Accept optional --title, --author, --description, --keywords / --tags, --publisher, --rights, --language, and --identifier arguments; expose editable fields in the web command builders. Unspecified values remain optional.
• Add regression tests for original software preservation, descriptive metadata mapping, explicit tags, and command generation; update README and OCR workflow documentation.

Metadata crosswalk

PDF	XMP	Meaning
/Author	dc:creator	Person or organization responsible for content
/Creator	xmp:CreatorTool	Original authoring software
/Producer	pdf:Producer	PDF-generating/enhancing software
/Title	dc:title	Title
/Subject	dc:description	Summary or description
/Keywords	pdf:Keywords, dc:subject	Keyword tags
—	dc:publisher, dc:rights, dc:identifier, dc:language	Additional descriptive fields
/Lang	dc:language	Document language

Reference: https://developer.adobe.com/xmp/docs/xmp-namespaces/dc/

Validation

GitHub's branch comparison confirms 10 modified files relative to main, and new test cases were added. Automated tests have not been executed in this environment because the Python PDF dependencies aren't available in the working runtime.