MSVC seems to have the best funding model of the three major compilers. I estimate that has more to do with adoption velocity than head starts on implementation, especially given how long modules have been in progress.
Clang had another model that was rejected, and most of the years in committee were actually about finding a compromise between Microsoft's and Clang's approach. In the end, Clang wasn't particularly happy about the end result either.
But integration and tooling is required for any major amount of adoption. IntelliSense has to work. Existing LSPs have to work.
That’s not a criticism necessarily , but just to add that I think they likely have a more agile code base as a result.