Read transliteration rules from an orthography profile

[xrotwang]

transliterate could support reading transliteration rules from an orthography profile (see Moran and Cysouw 2018) like the python package segments does.

[dwhieb]

I haven't had a chance to read the Unicode Cookbook yet, so I wasn't aware of orthography profiles. They look very promising! I'll spend some time reading about Moran & Cysouw's approach and then think about how this library might align with it better, as well as support importing transliteration rules like you're suggesting. Thanks!