Fix a non-existent macro unset by GHC

__wasm__ is a clang builtin macro when targetting wasm32, but GHC
passes -undef when invoking clang, therefore we can't use it. Use
wasm32_HOST_ARCH instead, since it's generated by the GHC build
system.

Fix compilation for wasm32-wasi

Bump bounds to accomodate base-4.18

GHC tests: mark process tests as broken for JS

docs: Add docs about OpenExtHandle in ProcessHandle__

testsuite: Don't attempt to read from inherited fd in processT251

It turns out that this test is subtly broken. In particular, the
test will fail if any file is opened in the subprocess before the child
is run since the closed fd 0 may be reused for the new file.
This tends to happen in the threaded RTS due to the event manager's
control pipe (see GHC #22395). Unfortunately, it's not really clear how
else #251 can reliably be tested.

Allow unix-2.8

Support `unix ^>=2.8`

Fixes: #258
See: https://github.com/haskell/process/issues/258

Fix process#251

Previously to spawn a process with a closed standard handle, we
would use `posix_spawn_file_action_addclose`. However, it turns out that
POSIX specifies that `spawnp()` may fail if `addclose()` is used on an
fd that is already closed. While glibc and musl appear to ignore this
aspect of the specification, Darwin indeed follows it leading to #251.

This behavior is rather unfortunate as
`posix_spawn_file_action_addclose` is a convenient way to close a handle
in a subprocess in a race-free manner (e.g. unlike `O_CLOEXEC`, which is
global). To avoid #251 we must first use
`posix_spawn_file_action_addopen` on the fd (e.g. opening `/dev/null`)
to be closed to ensure that it is valid, which has the side-effect of
closing the inherited fd. We can then safely use
`posix_spawn_file_action_addclose` to close the fd.

Fixes #251.

Darwin violates POSIX by making
`dup2(x,x)`, which should be a no-op, error. Consequently, we must take
care not to `dup2` in such cases.

We had already made this change in the `posix_spawnp` codepath but I had
assumed that this *only* affected `posix_spawnp`, not the `dup2` system
call itself.

winio: match mio in setting default encoding for pipes instead

winio: Fix Handle mode on inherited handles.

Allow Win32-2.13

fix cross-compiling configure issue

Bump base upper bound

testsuite/process011: Don't rely on Python

The refactor of `process011` performed in 8cd7e043, where
the test is taught to use `python3` instead of `sh`, has broken the test
on Debian 10. Specifically, for reasons that I don't yet understand,
`python3` exits with code 1 not code 2 when killing itself with
signal 2. Strangely, my tests on other distributions (e.g. NixOS)
suggest that Python's behavior in this case is rather inconsistent.

To avoid this, we now rather use a dedicated C program instead of
Python.

Fixes #241.