Yeah that's basically why I didn't pull it out as an option in the first place, it's not always practical. A lot of your proprietary code is going to be external dependencies linked/built against, or your own IP reused from the last project. But not all of it, and I can definitely see that smaller chunk causing a lot of problems.
You need a team that does a lot of dependency management and similar things well while building it, that don't actually help them get the game out faster, to keep the problem manageable. Or a team who specialize in open sourcing games like this, which could become a thing if this was more commonplace.
It's not that there wasn't any political pressure. It's that the slightest bit of pressure caused them to pull the plug swiftly.
I think the companies who were led by people personally antagonistic to DEI already weren't doing it. They started it when the political winds were in favor of DEI, found that it did something beneficial for them that was worth the investment (ultimately, increasing profits, probably through PR) and reaped what they could. But the slightest headwinds caused them to drop it, for lack of confidence it would be worth the continued investment. For others, it was beneficial enough this pressure didn't change their decisions.
None of this is likely coming from company leaders caring about DEI for some sort of principled reason, just companies who care about only one thing, reassessing the value of DEI in terms of that one thing, $ return on spend. This is a group who needs subtler treatment than the anti-DEI crowd, this is fair weather friends who don't care. What little we can do is reward those who don't give in to the slightest push.