Wrong image codes returned by occs/prevalence

[vjpsyverson]

As discussed here: Some of the image codes being returned by occs/prevalence are incorrect, specifically those for three of the subtaxa in the crinoid order Articulata. (They're showing pictures of articulate brachiopods instead, which is no longer an accepted name in the tree but probably still explains the error.) The problem is probably actually just incorrect entries in a table somewhere, but I don't know which one or how to fix it.