Too many protein columns selected when CDS columns without sequence variants are selected

1. Retrieve example Ensembl ID, select 'show cross references', then close the original window.
2. Open feature settings and limit display to CDS variants with clinical significance of pathogenic.
3. Right click variant feature row in feature settings - select 'hide columns that do not contain'.

This results in considerably fewer protein columns visible than expected. This is because the 'hide columns that do not contain' is applied to selected CDS columns not posessing visible variant sequence features, many of which map to an amino acid that also has a mapped variant. Nonetheless, the hide columns ensures those are also hidden.

In 2.11.1 onwards (with virtual features), this wrinkle complicates the procedure for showing just a specific set of protein positions mapping to one or more annotated CDS positions. For 2.11.2, the reccommended procedure is:
i. Select CDS columns containing features
ii. Use 'hide all but selected region' in *PROTEIN* view.

Ideally, this should be possible via the sequence features dialog instead.