Translate image OCR now uses File Processing

What changed

Translate image OCR now runs through the File Processing default image-to-text processor instead of the legacy OCR provider configuration. The default processor is System OCR on macOS and Windows, and Tesseract on Linux.

Why this matters to the user

Users who previously configured OCR providers for Translate will no longer see that legacy provider system. Translate follows the File Processing image-to-text default, so OCR behavior and processor selection now match the shared File Processing settings.

What the user should do

Review the default image-to-text processor in File Processing settings and choose a different processor there if needed.

Notes for release manager

This entry is specifically about the user-visible Translate OCR configuration change. The internal IPC cleanup is not user-visible.