v1.0.6

chore: drop pytesseract dependency, bump to 1.0.6

ocr.py no longer uses pytesseract — calls tesseract binary directly to
enable subprocess tracking and OMP_THREAD_LIMIT control.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

v1.0.5

chore: bump version to 1.0.5

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

v1.0.4

fix: resolve 4 bugs in TUI and compression pipeline

- CheckInputScreen.on_input_submitted: check_ocr() was running
  synchronously on Textual's main thread, freezing the UI on large
  PDFs; refactored to @work(thread=True) with loading state and
  disabled controls during the check
- _launch_ocr: replaced 3 hard-coded pop_screen() calls with
  _return_to_home() to avoid silent breakage if screen stack depth
  changes
- compress.py: removed /screen PDFSETTINGS for strength > 60 which
  caused a sudden JPEG quality cliff (30%) unrelated to DPI; all
  levels now use /ebook as base with quality controlled entirely by
  explicit DPI flags
- tui.py: fixed missing accent on "già" in OCR menu preview text

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

v1.0.3

chore: bump version to 1.0.3

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

v1.0.2

ci: remove smoke test step — .app bundle deleted after DMG creation

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

v1.0.1

build: add Applications symlink to DMG for drag-to-install UX

v1.0.0

ci: add contents:write permission for release upload