Add image OCR converter, document format updates, docs, and license

[etherlink-intern]

Motivation

• Add on-device image OCR support and make image inputs first-class in the conversion pipeline so apps can import photos and scanned images into Markdown.
• Document how to wire system import surfaces (file importer, PhotosPicker, Share Extension) into the library and record a plan for future native/FOSS backends (PDF/DOCX/PPTX/XLSX).

Description

• Introduce ImageOCRConverter that uses Apple Vision/CoreGraphics/ImageIO (guarded by #if canImport(...)) to extract recognized text lines and return a MarkdownDocument with OCR metadata.
• Extend DocumentFormat with image formats (png, jpeg, heic, tiff, gif) and add imageFormats helper; update format inference to recognize image extensions and MIME types.
• Include ImageOCRConverter in MarkItDown.defaultConverters so image inputs are routed to the OCR converter when available and otherwise result in unsupportedFormat when Vision is unavailable.
• Add comprehensive documentation Docs/ImportButtonIntegration.md for importing via document picker, PhotosPicker, and Share Extensions plus Docs/NativeConverterBackends.md which outlines the planned native/FOSS backends and integration policy.
• Update README.md to list supported input families and planned backends and add a project LICENSE (MIT).
• Add new/updated tests in Tests/SwiftMarkItDownTests that verify image format inference and conditional Vision OCR behavior, plus helper code to render OCR test images when platform frameworks are available.

Testing

• Ran the unit test suite with swift test on a macOS environment with Vision/CoreGraphics available, and the test suite (including the Vision OCR tests) passed.
• Verified conditional behavior on platforms without Vision by ensuring the image conversion path throws ConversionError.unsupportedFormat in unit tests, which passed under those configurations.
• Ran the CLI smoke tests via Scripts/smoke-test.sh and confirmed they completed successfully against the updated README and fixtures.

---

Codex Task