Add workaround for NCBI Datasets' incompatibility issues

[joverlee521]

Initially brought up by @tsibley in nextstrain/public#18 (comment)

Sigh. This isn't the first time this has happened (one example). Typically, new fields are backwards compatible with old software versions because the software is designed to be forwards compatible by ignoring fields it doesn't know about. NCBI's intentionally chosen to not include such forward compatibility, which means you basically need to stay in lockstep with the latest release. That's frustrating.

NCBI mentioned a workaround we could maybe use (№ 3):

Use datasets summary virus genome taxon 3048448 --as-json-lines | dataformat tsv virus-genome. This avoids triggering the error because datasets summary does not pick up the new field.

although I don't know how much overhead it adds compared to our current approach of downloading the bulk data package.

[joverlee521]

Hmm, we would essentially be downloading the metadata twice.

1. datasets download virus genome taxon - this would be for downloading the genomic sequences, I did not find way to exclude the metadata or just download sequences.
2. datasets summary virus genome taxon 3048448 --as-json-lines | dataformat tsv virus-genome - download and format metadata to avoid new field incompatibility issues.

This might be fine for small datasets like zika, but probably not ideal for large datasets like ncov or mpox.