`std::thread::available_parallelism` should respect process affinity and job object limitations

[Fulgen301]

Currently, the Windows implementation of std::thread::available_parallelism just returns the number of CPUs on the system:

rust/library/std/src/sys/pal/windows/thread.rs

Lines 126 to 136 in e3fccdd

	pub fn available_parallelism() -> io::Result<NonZero<usize>> {
	let res = unsafe {
	let mut sysinfo: c::SYSTEM_INFO = crate::mem::zeroed();
	c::GetSystemInfo(&mut sysinfo);
	sysinfo.dwNumberOfProcessors as usize
	};
	match res {
	0 => Err(io::Error::UNKNOWN_THREAD_COUNT),
	cpus => Ok(unsafe { NonZero::new_unchecked(cpus) }),
	}
	}

This is in stark contrast to the Linux implementation, which not only checks process affinity, but also queries cgroup information. It'd be better if the Windows implementation also respected process affinity and job object limitations.

[workingjubilee]

As I recall, these are only "best effort" things in each case i.e. the API already explicitly allows accounting for these or not, so it should be acceptable to try to account for this and simply ignore things if we get a confusing response from the relevant calls.