Hmm.
For the early titles listed, when the games came out, Linux was pretty irrelevant from a gaming standpoint.
Later, many games that had cross-platform releases used engines that provided cross-platform compatibility. Those games would have been written to the platform, so I'm sure that ports weren't as easy.
Now, the games are very elderly. The original team will be long gone. I don't know if there's anyone working on those at all -- unless a game represents some kind of continued revenue stream, there isn't a lot of reason to keep engineers on a game.
WINE runs them fine, so there's a limited return for Blizzard to do a native port. In fact, as I recall, Starcraft was one of the first notable games that WINE ran...I remember Starcraft support being a big deal around 2001, IIRC. The original Warcraft was for DOS, so you can run that in a DOS emulator.
I doubt that the investment in a Linux-native port in 2025 is going to get much of a return relative to what other things one could do with the same resources.
I guess maybe I could see an argument for World of Warcraft, as a very successful, long-running MMORPG that still has players and still represents revenue. But I think that I'd be surprised to see native ports of most of their earlier library.
Same reason that it's uncommon for any page to have most of the page covered in ink, regardless of whether it's a book or a sheaf of papers or whatever. Ink costs something, and it's cheaper to put ink on a little bit of the page than it is to put ink on everything but a little bit of the page. Unless there's a compelling reason to do otherwise, you take the cheaper route.