[FEATURE REQUEST] Integrate YomiToku OCR Engine into OSS Document Scanner

Is your feature request related to a problem? Please describe.
From a Japanese user’s perspective, the current OCR quality in OSS Document Scanner is not usable—recognition of Japanese text frequently fails, making the app impractical for day-to-day use in Japan.

Describe the solution you'd like
Please consider adding YomiToku as an optional (or default) OCR backend. YomiToku is an open-source OCR engine that has become popular in Japan over the past six months due to its high accuracy on Japanese documents. It is released under the CC BY-NC-SA 4.0 license, so it can be integrated into OSS Document Scanner for non-commercial use with proper attribution and share-alike compliance.

Describe alternatives you've considered

• Continuing to rely on the current generic OCR engine (insufficient accuracy for Japanese).

Additional context

• YomiToku GitHub: https://github.com/kotaro-kinoshita/yomitoku
• License: CC BY-NC-SA 4.0

No implementation files, tests, or entry points are named. Start by locating the current OCR backend in OSS Document Scanner, then review YomiToku and its CC BY-NC-SA 4.0 terms. Done would mean a documented optional or default YomiToku backend that improves Japanese recognition while meeting attribution and share-alike requirements.