It could be potentially exponentially more work to add such a feature in, versus how many people want it (and some people would want it disabled, which means that whether it was enabled or disabled would need to be accounted for at the end of every relevant scene, which to be honest would probably not be too hard to do but would likely take a long time. Then based on how the model works, they would need to change based on whether or not they're pregnant, and they would also need to add that for every existing model and outfit, which sounds like a lot of work and sounds like it would take exponentially more time for each outfit that existed in the game) And then if there are any errors that arrise between scripts and rendering, those would then need to be fixed, and every relevant animation would need to be checked to ensure it isn't broken by the model change. Every relevant scripted scene may also need to logically change with potentially different models exclusively for that scene. I wonder whether the dev would do it based on that alone.
It could made into a bullet point list, but I think the question would come down to how many people would really want it for whether the development time for it is worth it, and not better spent on another area. There's also the thing where it's potentially going through this whole process again each and every time a new romance character is added. And whether most people would turn it off, why even implement it?
My point is basically, it would likely depend on how many people would use the feature, it is worth the technical debt it may incur? By that I mean, potentially making it more difficult to add new features without a potential rewrite of the whole system. I don't know, but I could likely see the dev not doing it because of that.