Manage controlled vocabularies

[ar-jan]

Handle processing Lexicon of Linguistics lemmas for use in the annotation form.

This repeats each term per individual subfield (linguistic level) instead listing them under combined Phonetics/Phonology etc. groupings. Fixes the sort order for use in the form (order by the visible term instead of the lemma).

Not ready for merge: I'd like to see if we can fetch the list for use in the annotation form directly from SQLite. If so, we can add a separate TSV for Clarin concepts, create a SQLite database with both Lexicon and Clarin terms in the build step, and do one query to get a combined Lexicon+Clarin list.

[ar-jan]

First run python vocabulary/vocabulary.py to get the new JSON file. Can maybe done in a Github Action when building the .xpi.

[tijmenbaarda]

Looks good!