Reassess preventing "intermediate" or "unassigned" taxon being used

[leehart]

Re: prep_samples_for_cohort_grouping function

I know this is unchanged but I am not sure why we would prohibit anyone from using an "intermediate" or "unassigned" cohort by default. In almost all cases, there will not be enough samples to form a cohort (and if there ever are enough "intermediate" or "unassigned" samples to form a cohort using the default cohort_min_size, we should probably have a good think about what it means) but, in particular if someone manually labeled some samples as "intermediate" in a different column not named "taxon" for the specific purpose of studying them, they will be surprised for them not to show up, I think.

Originally posted by @jonbrenas in #694 (comment)

[leehart]

Needs further thought and discussion.