Dealing with lower case nucleotides: document and potentially auto-convert

[corneliusroemer]

Current Behavior

Our auspice.json schema expects nucleotides to be upper case and fails if mutations are lower case:

augur/augur/data/schema-export-v2.json

Lines 304 to 309 in 7cb3848

	"description": "nucleotide mutations",
	"type": "array",
	"items": {
	"type": "string",
	"pattern": "^[ATCGNYRWSKMDVHB-][0-9]+[ATCGNYRWSKMDVHB-]$"
	}

We don't seem to consistently validate this before export though and it can cause subtle bugs as our code often implicitly assumes nucleotides to be upper case.

It would be good to either validate this in various augur commands or do auto-conversion. I think auto-conversion is fine as long as it's documented.

I don't know exactly which tools use lower case but I have encountered lower case occasionally.

This is related to #1323 which does auto-conversion, but only in one place. There might be others that would benefit from auto-conversion.

Nextclade silently converts everything to upper case.