Entering diet ecology data changes basis for life habit

[mclapham]

We've been updating ecology information for some decapods and have run across some weird behavior that seems like it's probably a bug.

When updating diet/feeding mode, it forces the life_habit_basis to be at the same taxonomic level as the diet_basis. As a result, the life habit is blank unless something is entered at the same taxonomic level as the diet. This isn't ideal, because the whole point of the ecology system is that it inherits the trait from higher in the tree, and diet and life habit don't always correlate (i.e. we often want to set diet at one taxonomic level and life habit at another).

The reverse action, updating the life habit for a group, doesn't change diet_basis or any of the other fields. The motility_basis and vision_basis fields are not changed when entering diet, only life_habit_basis. We haven't tried updating motility or vision, so we don't know how or if those affect other fields. It seems like a bug because it's only editing diet that causes the change in life_habit_basis and apparently nothing else.