fix(stdio): launch PowerShell (.ps1) servers via a PowerShell host on Windows

[dltsum]

Summary

Fixes #3496

On Windows, get_windows_executable_command() resolves a stdio server command to a .ps1 path (via shutil.which extension probing), but CreateProcess cannot execute a PowerShell script directly — the spawn then fails with WinError 193 (%1 is not a valid Win32 application).

Fix: in create_windows_process, when the resolved command ends with .ps1, rewrite the argv to launch through a PowerShell host — pwsh preferred, powershell as fallback — with -NoProfile -NonInteractive -ExecutionPolicy Bypass -File <script> <args...>. If neither host is on PATH, raise a clear FileNotFoundError instead of the opaque WinError 193. Both the anyio.open_process path and the FallbackProcess (SelectorEventLoop) path benefit, since the rewrite happens before either spawn.

Testing:

• New Windows-only integration test test_a_powershell_script_server_round_trips_messages: writes a real .ps1 echo server, connects via stdio_client, and round-trips a JSON-RPC ping.
• Verified on Windows 11: fails without the fix (OSError: [WinError 193], exactly the reported error), passes with it (0.68s).
• ruff check / ruff format clean.

[github-actions]

This PR has been closed automatically. This repo only keeps pull requests open when they come from a maintainer, or from a contributor a maintainer has assigned to the linked issue, and you aren't currently assigned to #3496.

If a maintainer assigns you to #3496, this PR reopens on its own and there's nothing more you need to do here. Assignment is a maintainer call based on capacity; comments that only ask to be assigned don't factor in. What does help is engaging on the issue itself by confirming the repro, explaining why it matters for your use case, or describing the approach you'd take.

You're welcome to keep pushing commits here (just avoid force-pushing, since GitHub can't reopen a rewritten branch), but that on its own won't get the PR reviewed or the issue assigned, and realistically most auto-closed PRs stay closed. There's no need to open a new PR either way.

CONTRIBUTING.md has the full reasoning, but in short:

• We're a small team with very little capacity to review community PRs right now.
• Many recent PRs are AI-generated with little human review, and reviewing one carefully still costs a maintainer as much time as it ever did. A well-described issue is usually more useful to us than the code.

Maintainers: reopen, remove missing-issue-link, or add bypass-issue-check to override.