I apologize if what im suggesting might be off topic, but if such a feature did become a thing, why not a similar feature in the editor to force a specific type of lemming in the level and the set's default is what the $LEMMINGS line of text calls for what lemming to use in the set? and a drop down menu gives you the choice of what custom lemming sprite sheet you can use?
One of these is "add a checkbox to the options menu, that controls another "yes or no" setting - something that there's already many of, so it's written so it's easy to add / remove them. Then, if this option is on, ignore any lemming spritesets specified in the theme file, and just use the default lemmings instead". (EDIT: And ignore any recoloring data in the theme file, except maybe "MASK".)
The other is "add a new configuration item to the editor, which in turn needs to figure out a full list of which styles have lemming sprites and which don't; it additionally needs to support a "use the theme's specified set" option. If we're allowing equivalents to the theme-based recoloring, I then further need to implement an entire UI for this, which is also one of those "very easy to abuse" features - and then enter into the discussion (and add support, if decided) for overriding the other theme colors too, which the fact that all this exists would be seen as a strong argument towards "why not allow the rest". Or if recoloring the sprites is not allowed, I have to deal with the whole inevitable debate about
why not. Then, I need to figure out how to best represent this in a level file's format, and add that to loading and saving on both the editor and player side. Then, I need to implement actually applying this in the player, where currently, level data and lemming sprite data are more or less the programming equivalent of "entirely unaware that each other even exist". Then, on top of all that, I have to go and update the level format topic to reflect these new additions, and describe it in a way that's easy to understand." And all of this is for an option that will very rarely see use, and a significant proportion of the use cases for it are either "do it just because I
can" or "do it for trolling purposes".
So yeah - basically, one of these is a quick 5 minute option, the other is a
huge major feature. That's "why not", or more accurately, why "if we have the former, why can't we have the latter" is not really a meaningful point. As a suggestion in and of itself - I don't think it's very likely to happen, simply because it provides zero benefit to puzzles and isn't likely to see frequent use (plus, the abuse potential), but my biggest point here is that it's not comparable to this topic's request in scale, so there's no logic by which having one raises a "why not?" about the other.