Disable tests that consistently fail in the VM queues for performance-ci runs.

[LoopedBard3]

Disable tests that consistently fail in the VM queues for performance-ci runs, but keep them for internal runs.
This should clean up the last of the consistently failing tests in our performance-ci, where these tests are failing on the Open helix queues but are successful internally.

[Copilot]

Pull Request Overview

Modifies performance test configurations to disable flaky scenarios in VM queues for performance-ci runs while maintaining them for internal runs. The changes improve CI reliability by excluding tests that consistently fail in VM environments due to resource constraints or flakiness.

• Added conditional gating for three performance scenarios based on team project type
• Introduced IncludeVMFlakyScenarios property as an override mechanism
• Added explanatory comments documenting the reasoning for each exclusion

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
eng/performance/sdk_scenarios.proj	Gates Inner Loop Blazor Client scenario to internal runs due to higher resource usage
eng/performance/scenarios.proj	Gates Static Winforms Template and WinForms Large scenarios to internal runs due to VM flakiness

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[DrewScoggins]

Failures seem unrelated. Not sure if we want to go ahead and merge.

[xtqqczze]

These tests are not consistently failing, all tests were successful in https://github.com/dotnet/performance/pull/4998/checks?check_run_id=51930408911.

[LoopedBard3]

Failures seem unrelated. Not sure if we want to go ahead and merge.

The failures are unrelated, but I am doing a rerun to see if it is just flaky.

These tests are not consistently failing, all tests were successful in https://github.com/dotnet/performance/pull/4998/checks?check_run_id=51930408911.

While not failing on every single test run they are failing on a very large portion of PRs keeping us from being able to set auto-complete and needing to override to merge, so I think this is still a worthwhile update especially as there are still a lot of tests for coverage of these scenarios.

[DrewScoggins]

LGTM