filter / identify in OGS evidence and refined annotations where we have stop codons within gene

Artemis makes it easy to view stop codons. With Apollo, you need to be zoomed in all the way and these aren't really flagged as an error.

Some of this could be done within JB2 over the OGS by default. In the same way we have non-canonical splice sites we could detect or mark in-frame stop codons for an annotation within the refined annotations.