[mono][wasm] intrinsics improvements

[lewing]

Evidence suggests that there is value in filling some more of the Vector and Unsafe intrinsics

• Add intrinsics for 128 bit vector types
  • Use PackedSimd methods where they correspond directly to 128 bit Vector methods
  • Use PackedSimd methods for Vector(128)? operators for direct matches
  • add implementations of .NET missing intrinsics using Wasm intrinsics directly
• Implement managed PackedSimd implementations of the Vector128 methods that have managed sse implementations
• recognize VectorMath hot paths
• implement missing instance intrinsic methods
• implement missing Unsafe intrinsics
  • Bitcast
    • AOT [mono][aot] optimize aot Unsafe.BitCast for more cases #115130
    • Interpreter
  • ReadUnaligned, WriteUnaligned

PRs

• [wasm][aot] Optimize 64 bit const shuffles. Otherwise prefer vector swizzle. #115351
• [wasm][jiterp] Implement i64 jiterp shuffle and simplify logic #115344
• [wasm][aot] Fix op_OnesComplement on wasm and enable ceil/floor #115151
• [mono][aot] optimize aot Unsafe.BitCast for more cases #115130
• [wasm] Use packedsimd in hexconverter #115062
• [wasm] Try out a PackedSimd WideningLoader #114811
• [wasm][interp] Add Store intrinsics #114711
• [wasm][interp] Jiterp packedsimd shifts #114669
• [wasm][interp] Disable PackedSimd operator aliases while we investigate #114659
• [mono][interp] Avoid doing extra lookups when not needed #114517
• [mono][interp] Add convert aliases #114501
• [mono][interp] Add Sqrt packedsimd alias #114493
• [wasm][interp] Filter out overmatched shift #114492
• [wasm][interp] Fix PackedSimd.Splat intrinsic lookup #114413
• [wasm][interp] alias Vector operators to PackedSimd #114293
• [mono][interp] Group all the System.Runtime.Intrinsics logic together #114270
• [wasm][interp] Add aliases for Vector128.Load and Vector128.LoadUnsafe #114243
• [mono][interpreter] Make nint and nuint work for interpreter simd intrinsics #114087
• [wasm][interp] Match more Vector intrinsics to PackedSimd intrinsics #114070
• [wasm][interp] Add Round alias #114399
• [wasm][interp] Add the opcodes to the widening intrinsics #114395
• [mono][interp] Check packedsimd first on browser #113743
• [mono][interp] Handle missing Vector(128)?<double>.One case in interp #113739
• [mono][interp] Add SN Vector intrinsics #113456

9.0

• [release/9.0-staging][wasm][interpreter] Fix PackedSimd interpreter intrinsics #114218

[dotnet-policy-service]

Tagging subscribers to this area: @BrzVlad, @kotlarmilos
See info in area-owners.md if you want to be subscribed.

[dotnet-policy-service]

Tagging subscribers to 'arch-wasm': @lewing
See info in area-owners.md if you want to be subscribed.

[dotnet-policy-service]

Tagging subscribers to this area: @BrzVlad, @kotlarmilos
See info in area-owners.md if you want to be subscribed.

[dotnet-policy-service]

Tagging subscribers to this area: @steveisok, @vitek-karas
See info in area-owners.md if you want to be subscribed.

[lewing]

cc @kg