Arm64/Sve: Handle ConvertTo* APIs properly

[kunalspathak]

We were hitting an assert in these tests because we were not handling the wrapping of these intrinsics inside `ConditionalSelect.

• Make sure that ConvertTo* APIs get contained inside ConditionalSelect. To do that we now store the auxiliary type as baseType of return vector and if that matches with the wrapped mask operation, then we contain the intrinsics.
• Fix the codegen depending on the variant of the operation (see below)

ConvertToInt32 implements https://docsmirror.github.io/A64/2023-06/fcvtzu_z_p_z.html for following variants:

• Single-precision to 32-bit - FCVTZS <Zd>.S, <Pg>/M, <Zn>.S
• Double-precision to 32-bit - FCVTZS <Zd>.S, <Pg>/M, <Zn>.D

ConvertToUInt32 implements https://docsmirror.github.io/A64/2023-06/fcvtzs_z_p_z.html for following variants:

• Single-precision to 32-bit - FCVTZU <Zd>.S, <Pg>/M, <Zn>.S
• Double-precision to 32-bit - FCVTZU <Zd>.S, <Pg>/M, <Zn>.D

[dotnet-policy-service]

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

[kunalspathak]

@dotnet/arm64-contrib @amanasifkhalid PTAL

@ebepho - For ConvertTo*Int64, you need to follow along this path.

[amanasifkhalid]

LGTM, thanks for fixing this!

[amanasifkhalid]

Do you intend for this assert to go away once @ebepho implements ConvertToInt64, or should we keep checking for intrinsics with different base/return types? If the latter, I wonder if we ought to add a flag for these types of intrinsics.

[kunalspathak]

For now, we will update the assert, but may be later, we might consider adding a flag. For now, I want to see if there are more intrinsic categories that falls under this.