Add support for searching case names by acronym

[ProbablyFaiz]

This is tricky. Some thoughts in no particular order:

• We can parse the query such that an all-caps token is also searched as an acronym
  • For example, a search for NIFLA should be treated, roughly, as a search for N* I* F* L* A*, while also allowing for uncapitalized conjunctions, articles, and prepositions to not be included in the acronym.
• One way to do this would be to pass a prefix query just like the one above to Postgres with FOLLOWED BY (<->) operators in between, but Postgres doesn't seem to respect single letter prefix-matches. Look more into this.
• The other way to do this is to create a curated index as part of the case names to pre-generate possible abbreviations, and when we think there's an abbreviation, query that too.
  • Alternatively, just include the abbreviation(s) in the current searchable case index to have it get surfaced.

[ProbablyFaiz]

Look into Postgres FTS thesaurus/synonym dictionaries: https://www.postgresql.org/docs/9.1/textsearch-dictionaries.html