Add phonetic parsed type?

[jrdh]

Using phonetic algorithms (like soundex or metaphone) we can do string matching similar to fuzzy matching but based on pronounciation (often specifically English pronounciation). This may be worth adding as an additional field parsing type though testing would be required to ensure it was providing something fuzzy matching wasn't, or we could implement it but add it as a parsing option.

Libs like https://jamesturk.github.io/jellyfish/ implement a few different phonetic algorithms we could use, and in rust too so it's speedy. Need to ensure usefulness first though.