Initial version
This version is the result of experiments that have been going for a while.
It is not stable yet, but I am planning on removing several of the experimental scripts I wrote, mostly when I was validating the data from different collections such as ABC, TAVERN, Haydn Op. 20, and When-in-Rome.
I have a pretty standard workflow for pairing scores and annotations nowadays, so most of these scripts (e.g., measure_alignment.py) just clutter the repository.
Nothing major should have been omitted from this release.