Problem: users cannot abstract over tonal distinctions during search

[jrwdunham]

Problem

Users often request the ability to search their OLD forms while obliterating tonal distinctions. As a simplified example, there might be forms with transcriptions like tam, tám, tàm, tâm, etc. in the database. A user might want to search for tamand receive matches for all of the above. At present they could use regex to search fort[aáàâ]m` but that is laborious and requires some technical skill.

Acceptance Criteria

• Investigate the issue and come up with a way forward. Either implement it or create new sub-tickets if it is a significant piece of work.
• I am hoping we can use Unicode character analysis to distinguish "base" glyphs from "dependent" ones (TODO: correct terminology) and use this information to build some kind of general-purpose "tone destruction" functionality as a DativeRF search features. I am also hoping that we can use the existing advanced search functionality of the OLD as-is to support this.