generate present_in_taxon assertions for taxon union classes

[balhoff]

Taxon union classes are frequently confusing, because they generally are weaker statements than editors intend. For example, #29009. I think we can express better what editors intend if we couple the union expression with present_in_taxon for each term.

So, for Viridiplantae or Bacteria or Euglenozoa, if the term is also known to be only in Eukaryotes, we want to get a contradiction. We could, if this term also came with a present_in_taxon Bacteria statement. But need to think through how this works with the unions class hierarchy.