make musl cgroupv2 identifying same as other target_envs

[nrxus]

I've been noticing an issue where creating a cgroup instance with an auto hierarchy fails and the error seems to come from:

cgroups-rs/src/cgroup.rs

Lines 556 to 560 in 8d29c19

	if !fp.exists() {
	if let Err(e) = std::fs::create_dir(fp.clone()) {
	return Err(Error::with_cause(ErrorKind::FsError, e));
	}
	}

.I believe this is because cgroups-rs thinks that the machine is using cgroup v2 where it isn't. This is happening at a customer site so I can't be sure 100% but we do compile all of our binaries with musl as the target as we want static builds. The system is an older aws endpoint with a 4.14 kernel so I'd assume it's using a hybrid cgroup (where systemd makes a fake unified cgroup but it's still cgroup v1 in reality) instead of a pure cgroup v2 system so that may be where the musl specific cgroupv2 check fails. I think we can avoid this musl specifc code now that the crate is using v0.25 of nix as the needed constant was added in this version: nix-rust/nix@23f1787

This PR just removes the cfg gates so that musl targets hit the same code as other targets.

I also considered updating nix while i was at it since it is a couple of versions behind but I wasn't sure what your policy was for upgrading crates that cause an MSRV bump.

[Tim-Zhang]

@liubin I'm not familiar with this so please take a look at this.

[liubin]

@nrxus Now nix is 0.25.0, so it could compile.

[liubin]

Close and reopen to trigger the CI.

[Tim-Zhang]

@nrxus CI was failed, you can fix it follow the error message, thanks

[nrxus]

@Tim-Zhang I have fixed the commit message to match the expectations from CI (as per the contributing guidelines)

[Tim-Zhang]

LGTM, thanks @nrxus