Use the better pseudo-names where available on xarch

[tannergooding]

The vector comparison instructions all have preferred pseudo-names that make the disasm easier to read. We should prefer using them

[dotnet-policy-service]

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

[Copilot]

Pull Request Overview

This PR enhances disassembly readability by using preferred pseudo-names for various x86/x64 vector comparison and carry-less multiplication instructions.

• Introduces fltCmpOpNames and intCmpOpNames arrays for mapping immediates to mnemonic suffixes.
• Updates genInsDisplayName to format vcmp* and vpcmp* instructions using those pseudo-names.
• Adjusts emitDispIns to skip printing immediates for the newly supported instructions.

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File	Description
instr.cpp	Added pseudo-name arrays and switch handling for VEX/EVEX instructions in genInsDisplayName.
emitxarch.cpp	Modified PRINT_CONSTANT path to omit immediates for the new pseudo-named instructions.

Comments suppressed due to low confidence (1)

src/coreclr/jit/instr.cpp:121

• There don't appear to be unit tests verifying that each immediate value correctly maps to its pseudo-name; consider adding tests for a few representative values (e.g., control=0, 3, 7) for each instruction to ensure coverage.

    static const char* const fltCmpOpNames[] = {

[tannergooding]

CC. @dotnet/jit-contrib for review.

Some minor cleanup to the xarch disasm output to make vector comparisons more readable by using the friendly pseudo-names where available. As part of that, it refactored the general logic for displaying a constant from a bunch of goto PRINT_CONSTANT into a reusable helper method that utilizes the relevant instrDesc queries instead.