Github Instructions for onboarding performance related code and pipeline

[gearama]

Added copilot instructions for Rust performance framework onboarding.
The instructions are "burrowed" from Heath Stewart's PR about copilot instructions. I used those for consistency so when his PR is merged we work on similar assumptions and don't confuse the instructions
.
The supported operations are :
Onboard perf framework for package azure_security_keyvault_certificates.
Create perf pipeline for package azure_security_keyvault_certificates.

The code file generated might need some quick edits to make it work.

I would not say this is a definitive list. Improvements are welcome.

[Copilot]

Pull Request Overview

This PR introduces copilot instructions for onboarding a Rust performance framework and creating a perf pipeline for the package azure_security_keyvault_certificates.

• Added guidelines for setting up performance tests in Rust.
• Defined steps for configuring cargo.toml and creating YAML pipeline files.

Comments suppressed due to low confidence (2)

.github/copilot-instructions.md:12

• The path uses a backslash. Consider using a forward slash (benches/benchmark.rs) for consistency with common Rust directory conventions.

* Create a benches\benchmark.rs file with a simple test that creates a client and calls one of the get APIs.

.github/copilot-instructions.md:9

• [nitpick] The placeholder 'named name' may be confusing. Consider using a more explicit placeholder like 'package ' to clarify its intent.

## Onboard perf framework for package named name

[heaths]

Let's hold off. I want to get the initial copilot-instructions.md in first. How they are now isn't quite right. Then we can update this PR to add perf-related instructions; though, a few things I see off the bat:

1. Use more concrete wording. "Add performance support to crates" is going to work better, most likely, than using words like "onboard" or "perf". The clearer we are the better it'll work.
2. Too much written detail. Your example near the bottom is likely going to yield much better results.
3. After working in our Azure MCP code base a little, I wonder if this is too much to put in general instructions. That's why I actually moved commit and pull request instructions out. I wonder if we should actually consider using prompt file examples for specific use cases like perf tests and examples.

For the latter, I'd encourage you to give that a try. The more we put in copilot-instructions.ms, 1) the slower and more confusing responses we'll probably get, and 2) the number of tokens will cost more. We need to find that sweet spot, and adding new perf tests isn't exactly the highest priority order.