TLSH_PG: Implement an index to speed up searches

[rjzak]

After a few million records, searching for similar files becomes slow.
See: https://tlsh.org/papers.html

Ref: malwaredb/malwaredb-rs#165

[zmt-Eason]

Is this in progress?Do we have milestone for this performance?

[rjzak]

It's not yet in progress and I don't have any milestones yet. I've been reading up on Postgres internals but haven't started work on it yet.

I still need to understand the TLSH indexes properly and see how the Postgres indexing API works. I'm hoping to get some inspiration from the pg_vector project which does this.