Implement a --ignore-rna-in-tiering flag

[susannasiebert]

From #643 (comment):

There should probably also be a flag --ignore-rna-in-tiering(?) that allows for tiering to treat the experiment as if it has no RNA (even if some of those values exist) - this could be useful when the RNA is of really poor quality.