v1.6.0

[1.6.0] - 2026-07-01

Added

• Added comprehensive unit testing suite (tests/test_dictionary_builder.py) validating normalizers, morphology, and XML rendering rules.
• Added cross-platform auto-compiler detection supporting Windows Kindle Previewer paths.
• Added interactive near-match suggestions to CLI word preview using case-insensitive SQL LIKE queries.
• Added database post-build cleanup option (--cleanup CLI flag) to drop staging tables and run vacuuming.

Changed

• Refactored Bangla Morphology Engine to perform POS-aware conjugations (verbs end in ওয়া or standard া vs nominal endings).
• Enhanced English Morphology Engine with irregular rules and short adjective limits.
• Wrapped SQLite connection lifecycles in safe try...finally resource managers.
• Corrected MinhasKamal ingestion license metadata mapping from GPLv3 to MIT.
• Standardized logging under custom configuration writing to build.log and sys.stdout with ANSI filters.
• Sanitized MuntashirAkon raw markup (converts   to   and strips trailing unclosed tags) to guarantee well-formed XHTML.

Fixed

• Fixed layout nesting schema to place <idx:infl> inside <idx:short> (outside of <idx:orth>).
• Added XML attribute quote-escaping for single and double quotes.
• Fixed case-sensitivity metadata checking in frontend website (docs/app.js).
• Propagated compilation failure exit codes in GitHub Actions headless runner.