feat: integrate Tesseract.js with improved language availability and font handling

- Refactored OCR page recognition to utilize a configured Tesseract worker.
- Added functions to manage font URLs and asset filenames based on language.
- Implemented language availability checks and error handling for unsupported languages.
- Enhanced PDF workflow to display available OCR languages and handle user selections.
- Introduced utility functions for resolving Tesseract asset configurations.
- Added tests for OCR functionality, font loading, and Tesseract runtime behavior.
- Updated global types to include environment variables for Tesseract and font configurations.

docs/index.md

@@ -32,5 +32,11 @@ features:
     details: Convert, edit, merge, split, compress, sign, OCR, and more. Everything you need in one place.
   - icon: 🌐
     title: Self-Hostable
-    details: Deploy on your own infrastructure. Docker, Vercel, Netlify, AWS, or any static hosting.
+    details: Deploy on your own infrastructure. Docker, Vercel, Netlify, AWS, or fully air-gapped environments with self-hosted OCR workers, language data, and text-layer fonts.
+
+## Offline OCR
+
+If you self-host BentoPDF in an air-gapped or offline environment, OCR needs more than the Tesseract worker and traineddata files. Searchable PDF output also needs the OCR text-layer fonts to be served internally.
+
+See [Self-Hosting](/self-hosting/) for the full setup, including `VITE_OCR_FONT_BASE_URL`, the bundled `ocr-fonts/` directory, and the updated air-gap workflow.
 ---