gnutls: Fix failing patch for Aarch32

[matthewmazzanti]

Motivation for this change

A patch for tests that fail on arm has been upstreamed, and is no longer
needed.

Things done

I have gated this behind a lib.versionOlder to allow the build.

• Tested using sandboxing (nix.useSandbox on NixOS, or option sandbox in nix.conf on non-NixOS linux)
• Built on platform(s)
  • NixOS
  • macOS
  • other Linux distributions
• Tested via one or more NixOS test(s) if existing and applicable for the change (look inside nixos/tests)
• Tested compilation of all pkgs that depend on this change using nix-shell -p nixpkgs-review --run "nixpkgs-review wip"
• Tested execution of all binary files (usually in ./result/bin/)
• Determined the impact on package closure size (by running nix path-info -S before and after)
• Ensured that relevant documentation is up to date
• Fits CONTRIBUTING.md.

[ajs124]

Why add the versionOlder and not just drop the patch?

[matthewmazzanti]

In the same file theres a "versionAtLeast" here, so I wanted to keep with the flow and maintain the ability to change versions if needed. If it would be more appropriate to just drop the patch, let me know.

[ajs124]

We only tend to do that for packages with a common infrastructure that multiple versions are built from. Apparently this used to be the case for gnutls until 33e7d72/#58611.

IMHO having conditions which only ever evaluate to the same result is an antipattern and should be avoided, but if it is to be expected that a generic.nix and multiple different gnutls versions will return, it might make sense to keep them around anyways.

[matthewmazzanti]

Well I can remove the patch entirely, thats no problem. Do you want me to remove the other version dependent reference as well, or leave it in?

[matthewmazzanti]

Removed the patch.

[SuperSandro2000]

Can we target master here?

[matthewmazzanti]

Tried to change the base directly and it looked nasty. Let me do a cherry pick and open a new PR.

[matthewmazzanti]

New pr here: #117279