Fix SupportsSettingZeroFlag for shift/rotate nodes

[EgorBo]

Fixes #116125

Introduced in #113037

Zero diffs

[Copilot]

Pull Request Overview

This PR fixes a JIT compilation bug where shift and rotate operations incorrectly reported that they support setting the zero flag in all cases. The fix ensures that shift/rotate instructions only claim to support setting the zero flag when the shift count is guaranteed to be non-zero, since x86 shift/rotate instructions don't update flags when the count is zero.

• Adds proper condition checking for shift/rotate operations in SupportsSettingZeroFlag()
• Includes a regression test that reproduces the original issue
• Separates shift/rotate logic from other arithmetic operations that unconditionally support zero flag setting

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

File	Description
src/coreclr/jit/gentree.cpp	Updated SupportsSettingZeroFlag() to check if shift count is non-zero for shift/rotate operations
src/tests/JIT/Regression/JitBlue/Runtime_116125/Runtime_116125.cs	Added regression test with division by shift result that triggers the bug
src/tests/JIT/Regression/JitBlue/Runtime_116125/Runtime_116125.csproj	Test project configuration with optimization enabled

[dotnet-policy-service]

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

[EgorBo]

PTAL @jakobbotsch @dotnet/jit-contrib simple change - I've checked it with Intel and AMD manuals. Also, native compilers do similar checks.