variant data pipeline: replicate any metadata cleanup steps done by Bedford group

[elray1]

we think there might be stuff about things like location cleanup. it'd be good to make sure we do the same things as they do.

[bsweger]

No longer required--we're now pulling the metadata from Nextstrain instead of NCBI.