Add Vision-based Image OCR converter, docs, and tests

[etherlink-intern]

Motivation

• Add on-device OCR support so image inputs can be converted to Markdown using Apple Vision when available.
• Wire image-format recognition into the format inference so image files are routed to an OCR converter instead of returning unsupportedFormat by default.
• Provide example app wiring for an Import button so apps can easily hand picked documents or photos to the converter.

Description

• Add ImageOCRConverter which uses conditional imports for Vision, CoreGraphics, and ImageIO to run text recognition and return a MarkdownDocument, and otherwise throws ConversionError.unsupportedFormat.
• Extend DocumentFormat with image kinds (png, jpeg, heic, tiff, gif), add imageFormats set, and map common content-types and extensions.
• Register ImageOCRConverter in MarkItDown.defaultConverters so images are handled by the pipeline when supported.
• Add Docs/ImportButtonIntegration.md with SwiftUI fileImporter/PhotosPicker examples and update README.md; expand tests in SwiftMarkItDownTests to cover image format inference and conditional OCR rendering tests.

Testing

• Ran the unit test suite with swift test, which executed SwiftMarkItDownTests (including conditional OCR tests) and completed successfully.
• The tests include platform-conditional OCR rendering tests that exercise Vision-based recognition when the test host provides Vision/CoreGraphics/ImageIO, and they assert the expected unsupportedFormat behavior when those frameworks are unavailable.
• Verified repository-level CLI smoke tests with Scripts/smoke-test.sh to ensure basic CLI scenarios remain unchanged and they passed.

---

Codex Task