Apple has practically zero presence in enterprise (where one company can have 60,000 computers), and also practically zero in SMB.

Right, that's because since, well, NeXT acquisition they very openly abandoned that whole area and turned to marketing their stuff as home stuff and fashionable toys. That may change in future, like everything else.

Business software is written for windows. Even trying to use a Mac with the most basic office software is challenging - even if the exact same product exists in both.

The funny part is - Apple doesn't need a full transition to be an option. They just need to make their system more and more usable (cutting cost and cutting their usual bullshit too, so maybe not too soon) for similar things over time. Then some small businesses may start using it, then bigger ones, maybe also in niche roles (like it is even now with audio production and publishing, I think? not sure, I'm not an Apple user).

Legal - they all use a small set of document apps (which until recently was wordperfect), and some legal database apps. None of the database apps run on Mac as far as I’ve seen.

Apple may participate significantly in the Wine project to change this. They are a big company with resources.

Engineering - there are practically no CAD apps for Mac. Some do exist, but again, even the ones that are on both Windows and Mac are problematic at best on Mac, typically unable to integrate with the back end.

Apple also does have the weight to persuade the developers of mainstream CAD apps port them to MacOS. I don't think technical difficulties are the most important ones there. It's just that there were no reason to do it. Like no agreement, no common strategy, no deals. Apple wasn't interested in it because it's not their intended market, developers weren't interested in it because it's a small market.

Most people don’t have the bandwidth to learn a new system just to avoid the shitty part of Windows (which only affects home users anyway). It takes less effort/time to figure out how to mitigate the Windows issues than to deal with a completely new system, that will also have issues integrating with other stuff they already have.

There were differences in UX between any pair of a thing which lost popularity and a thing to which the former lost it. I'm not sure this is a good argument.

Also they can try playing the long game and expect more users in 5-10 years, not right now.