Refactoring of POSIX process logic

[bgamari]

This is a complete rewrite of posix/runProcess.c. There are a few goals
of this rewrite:

• fix a long-standing and serious bug in the execvpe fallback path, which
  uses non-reentrant functions after forking. This is of course undefined
  behavior and has been causing failures under Darwin's Rosetta binary
  translation engine (see GHC #19994).

• eliminate code duplication in the fork/exec implementation.

• introduce support for posix_spawn, allowing us to unload a significant
  amount of complexity in some cases. This is particularly desireable as the
  cost of fork has increased considerably in some cases on recent Darwin
  releases (namely when MAP_JIT mappings are used; see 1)

While posix_spawn is often a win, there are unfortunately several cases where
it cannot be used:

• posix_spawn_file_actions_addchdir_np is broken on Darwin

• POSIX_SPAWN_SETSID is only supported on mac 10.15 and later, but doesn't
  return a proper error code when not supported

• the originally-specified semantics of posix_spawn_file_actions_adddup2 are
  unsafe and have been amended (see 3) but not all implementations have
  caught up (musl has 4, glibc did later 5, Darwin seemingly hasn't) there appears
  to be no support at all for setuid and setgid

• spawn is significantly slower than fork on some Darwin releases (see 6)

To address this we first try using posix_spawn, falling back on fork/exec
if we encounter a case which the former cannot handle.

[snoyberg]

This is a large enough change in code that I'm honestly not that familiar that I can't offer a meaningful review. With tests passing, and based on you as the author, I feel comfortable merging it. I'd probably also pull this into typed-process and run the test suite there as well. I've added a suggestion to fix the CI failure, and then I'll do a little more testing.

[bgamari]

Hi Michael, I believe this is now ready for further testing if you are able. Otherwise, I have also done a fair amount of testing locally (on Darwin and Linux) and I believe the work is sound.

[snoyberg]

I've done additional testing on both Windows and Linux, by running both the local test suite and the typed-process test suite. Both are now passing. Merging now, thanks! Do you want me to make a release at this point?

[bgamari]

Thanks, @snoyberg! I've opened #209 with a small fix and a minor refactor. Otherwise I think we are ready for a release. I think we will use this release in 8.10.6, 9.0.2, and 9.2.1

[snoyberg]

Cool, release 1.6.13.0 made including the changes from #209.

[Bodigrim]

Builds fail for 1.6.13.0 with

cbits/posix/find_executable.c:12:20: error:
     fatal error: common.h: No such file or directory
   |
12 | #include "common.h"

https://matrix.hackage.haskell.org/#/package/process/1.6.13.0