I couldn't figure out the best area label to add to this PR. If you have write-permissions please help me learn by adding exactly one area label.

Looks good, I will consolidate the remaining changes from #63357 when you merge this.

Now that we skip that test we make it to Vector128_ro which likewise times out:

https://dev.azure.com/dnceng/public/_build/results?buildId=1586485&view=ms.vss-test-web.build-test-results-tab

@echesakovMSFT It might be better/easier to just skip these tests on Arm32 exclusively. Or maybe to add a special fast path in the tests (they are autogenerated so this would be easy, just a lot of churn) that effectively does an if platform == arm32 { early exit }

If the issue here is indeed that the combination of process startup + throwing exception is too slow on Arm32 (I imagine x86 Linux is also impacted), the most trivial fix might be to just assume that IsNotSupported is correct since we aren't really updating those platforms at all.

The NotSupported logic then at least gets validated on the x64/Arm64 Outerloop when we test COMPlus_EnableHWIntrinsic=0