Test Out New AssemblyAI Model (Universal 3.5-Pro)

## Summary

The new AssemblyAI model supports up to 30 speakers with speaker diarization (up from our original ~10).

This is a substantial lift in terms of our ability to potentially map voice lines with speakers in our hearing transcripts - we should test out the new model to determine:
* Does the speaker diarization now work properly (First level: Does it distinguish the speakers correctly? Second level: Can it name the legislators (whose names are known via committee/agenda)?)
* We can provide a list of known names to help improve accuracy - we should send in the list of all legislators on the Committee holding the hearing at minimum (possibly there's more specific data in the agenda)
* Cost Difference (if any)
* Compare to the `transcriptions//utterances` to see how different the diarization actually is between versions?
*

We would have to re-process all of our hearings with the new model to get this if it proves useful - so we should determine how much of a lift it really is in terms of data quality.

Start by reviewing the existing AssemblyAI transcription flow and the transcriptions//utterances data. Test Universal 3.5-Pro on a hearing using committee or agenda names, then compare speaker diarization, legislator naming, and cost with the current results. Done means documenting the data-quality and cost differences well enough to decide whether reprocessing hearings is worthwhile.

Written by the indexing model from the issue text.