Inconsistency in mis-spelled opinions; can not download some mis-spelled taxa

[pnovack-gottshall]

I've noticed an unusual behavior in the PBDB with regard to mis-spelled taxon names. It seems that opinions that parent a mis-spelling to some higher taxon get returned in downloads (and are available in the search bar), but that opinions that correct a mis-spelling (but where the parent is the correctly spelled name instead of a higher taxon) are not available as obsolete names in downloads/search bar. The mis-spelled names seem to always be printed on taxon pages, however.

Here's an example:
Sepkoski had 3 versions of the trilobite genus Ascetopeltis (Ascetopeltis, Ascopeltis, and Acetopeltis). Here are the opinions:

If you download all names for Ascetopeltis : https://paleobiodb.org/data1.2/taxa/list.csv?datainfo&rowcount&base_name=Ascetopeltis&rank=genus&variant=all

you only get the names for Ascetopeltis (even if only search for invalid or obsolete names, and even if search for all opinions).

Similar, if you download the mis-spelled name/opinion: https://paleobiodb.org/data1.2/taxa/list.csv?datainfo&rowcount&base_name=Ascopeltis&rank=genus&variant=all

you return nothing.

In contrast, if a mis-spelled opinion was entered with a higher-taxon parent, such as Rhytidopilus:

Then the mis-spelled name gets returned in the download, and is also available as a redirect in the taxon search bar:
https://paleobiodb.org/data1.2/taxa/list.csv?datainfo&rowcount&base_name=Rhytidopilus&rank=genus&variant=all

Is this the expected behavior for downloads? It seems inconsistent to me.

Alternatively, is there some way to consistently download ALL obsolete mis-spelling of names (e.g., Ascopeltis) that point them to their accepted names?