Replace use of target dependent TestZ intrinsic

[xtqqczze]

Contributes to #101251.

[dotnet-policy-service]

Tagging subscribers to this area: @dotnet/area-system-runtime-intrinsics
See info in area-owners.md if you want to be subscribed.

[xtqqczze]

Depends on #102705.

[xtqqczze]

@MihuBot

[xtqqczze]

@tannergooding There are unexpected regressions, see MihuBot/runtime-utils#501

System.Text.Ascii:IsValidCore[ubyte](byref,int):ubyte

 G_M58774_IG07:
        cmp      esi, 64
        jg       SHORT G_M58774_IG08
        vmovups  ymm0, ymmword ptr [rdi]
-       vpor     ymm0, ymm0, ymmword ptr [rax-0x20]
-       vmovups  ymm1, ymmword ptr [reloc @RWD32]
-       vptest   ymm0, ymm1
+       vmovups  ymm1, ymmword ptr [rax-0x20]
+       vmovups  ymm2, ymmword ptr [reloc @RWD32]
+       vpternlogd ymm0, ymm1, ymm2, -88
+       vptest   ymm0, ymm0
        sete     cl
        movzx    rcx, cl
        jmp      SHORT G_M58774_IG04
-						;; size=35 bbWeight=0.50 PerfScore 10.75
+						;; size=42 bbWeight=0.50 PerfScore 12.00

[tannergooding]

That particular case should be mostly resolved with #104517, vpternlog was missing general support for ensuring we select optimal containment

There's more improvements that could be had as well, but its a step in the right direction overall

[tannergooding]

It's not really a regression per say and in other cases could be an optimization.

Either way, it shouldn't be impactful to this PR, I'll just finish the remaining work here and not fold in the case the AND is part of an op_Equality check

[xtqqczze]

@tannergooding Depends on #104944.

[tannergooding]

I wouldn't say this depends on #10944, I think it's fine to take as is. If you could mark it ready for review then I can give it a final pass and merge

[xtqqczze]

@tannergooding I've suggested we make the suggested changes in a follow-up PR.