[codex] Use reading exposure in card ranking

[bee-san]

Summary

• Load optional generic Reading Exposure Exporter word media from Anki collection.media.
• Blend exposure recency/frequency into card ranking with configurable readingExposureWeight.
• Surface exposure score/count fields in sort previews and document the behavior.

Details

The sorter now reads _reading_exposure_words.json.gz from collection.media when present. _reading_exposure_manifest.json is supported for contract validation, but exporters should use the standard generic word filename rather than custom paths. The loader also falls back to legacy _kani_reading_exposure_words.json.gz for compatibility. Missing files are ignored; malformed existing files are reported in sort warnings.

The ranking signal favors recent exposure first, then 14/31-day windows, with a small lifetime-count component. It is additive to the existing frequency score and defaults to a modest weight of 0.18.

Validation

• python -m unittest discover -s tests -p 'test_*.py' -v (121 tests)
• python -m compileall addon/anki_sorter tests
• python scripts/package_addon.py
• git diff --check origin/main...HEAD and git diff --check
• Live read check against /home/bee/.local/share/Anki2/User 1/collection.media loaded 8,598 exposure entries with no warnings.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud