Add System.Diagnostics.Process output-reading benchmarks

[Copilot]

Extends Perf_Process with three benchmarks that measure reading standard output from a child process echoing 1000 lines.

New benchmarks:

• ReadOutputLineByLine — event-driven via OutputDataReceived + BeginOutputReadLine(); handler removed after exit
• ReadOutputToEndAsync — _ = await StandardOutput.ReadToEndAsync() + await process.WaitForExitAsync(); #if NET gated (not available on Full Framework)
• ReadOutputToEnd — synchronous _ = process.StandardOutput.ReadToEnd()

Shared setup:

• s_outputStartInfo: cmd /c for /L %i in (1,1,1000) do @echo Line %i on Windows, sh -c "for i in $(seq 1 1000); do echo Line $i; done" on Unix; RedirectStandardOutput=true, UseShellExecute=false
• s_ignoreOutputLine: static readonly DataReceivedEventHandler with a static lambda — avoids per-iteration delegate allocation and closure in ReadOutputLineByLine

Results are explicitly discarded via _ = to prevent potential dead code elimination. The async benchmark uses await process.WaitForExitAsync() to avoid mixing sync and async code.

---

📍 Connect Copilot coding agent with Jira, Azure Boards or Linear to delegate work to Copilot in one click without leaving your project management tool.

[adamsitnik]

@copilot please address my feedback

[Copilot]

@copilot please address my feedback

All four suggestions addressed in d3c0529: static lambda, _ = discards for both ReadToEnd calls, and await process.WaitForExitAsync() in the async benchmark.

[adamsitnik]

LGTM assuming the CI is going to be green