Utilized the version 2 pipeline for unified processing of TXT and OCR PDFs, and fixed dozens of bugs that could potentially lead to poor output quality.
This tag has no release notes.