[AmanJ24]

italian_60k and italian_280k had zero accented words (à è é ì ò ù), while the 7k list has 136. So the big lists were just damaged somewhere along the way, and the site code is fine. I only changed the two JSON files.

Three things were wrong:

• Words like perché, già, città, università were missing
• Broken stubs where the accent got deleted along with the last letter (tornò → torn, cioè → cio)
• Accent-less copies like perche, citta, piu

I removed the stubs and copies and added the accented words in their alphabetical spots. Everything else is untouched.

list	words	accented words	removed	added
italian_60k	60,442 → 64,536	0 → 6,506	2,412	6,506
italian_280k	279,833 → 292,122	0 → 18,108	5,819	18,108

For word data I took the accented word forms from Morph-it! and used the LibreOffice Italian dictionary and the FrequencyWords Italian list to check them. I also used the repo's own 7k list as a check and added its accented words. I think the licences are all GPL 3 compatible, but I'm not a lawyer, so tell me if you'd rather I use another source. I also filtered out vulgar verb forms from what I added.

Checks: pnpm check-assets-languages passes, formatting passes, no duplicates, and all 136 accented words from the 7k list are now in both big lists.

Heads up:

• The diff is big, so the numbers above are the easiest way to review it.
• I can't prove the lists are complete. Some accented words may still be missing, and a few bad entries may remain (like rag in the 60k, a stub of ragù). I checked by sampling.
• I didn't test it in the app, since it's data only.

Closes #4355