JIT: Added Sve.LoadVectorNonTemporal/NonFaulting/128AndReplicateToVector APIs

[TIHan]

Contributes to #99957

Adds:

• LoadVectorNonTemporal
• LoadVectorNonFaulting
• LoadVector128AndReplicateToVector

[ghost]

Note regarding the new-api-needs-documentation label:

This serves as a reminder for when your PR is modifying a ref *.cs file and adding/modifying public APIs, please make sure the API implementation in the src *.cs file is documented with triple slash comments, so the PR reviewers can sign off that change.

[ghost]

Note regarding the new-api-needs-documentation label:

This serves as a reminder for when your PR is modifying a ref *.cs file and adding/modifying public APIs, please make sure the API implementation in the src *.cs file is documented with triple slash comments, so the PR reviewers can sign off that change.

[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.

[TIHan]

sve_load_vector_results.txt

@dotnet/arm64-contrib @kunalspathak this is ready. The test results do have failures for NonTemporal and 128AndReplicateToVector because of predicate register support.

[a74nh]

I think this file is doing the same job as SveLoadMaskedUnOpTest.template, which is being used for
Vector<byte> LoadVector(Vector<byte> mask, byte* address) etc.

Your new file is better. Can you combine them both into a single template? (either naming is fine)

[kunalspathak]

+1

[TIHan]

Originally I wanted this, but it's a bit more difficult because the other Load APIs take different operand types. Can we do that one in another PR?

[kunalspathak]

Overall looks great. Added few questions and comments about test.

[kunalspathak]

+1

[kunalspathak]

LGTM. Thanks!