Deprecate `use_std` cargo feature: use `std` instead .

Related to #657 .

Fix locale_t in unix and fuchsia.

Closes #1055.

Closes #1055.

Run CMSG tests on s390x

Closes #1240.

Closes #1240.

Related to #657 .

Fix Android build jobs

Non-existent constants like SYS_syscalls have been removed

* Update Android NDK to version r19c
* Update Android API versions to:
  * API 24 on arm and aarch64 targets
  * API 28 on x86 and x86_64 targets
* Unified headers were removed in NDK 16
* Refactor the NDK and SDK installation scripts
* OpenJDK version 8 must be kept: it appears that
  the Android tools do not work with more modern OpenJDK
  versions.

openbsd: add KERN_PFSTATUS and update KERN_MAXID

`KERN_PFSTATUS` has been recently added, and so `KERN_MAX` value has changed.
Add it to avoid failure on (local) ci.

While here, I added few blank lines to separated series of constants.

Netlink constants for use with the route netlink subsystem

Reopening a PR to close #1059. I'll be adding a few more constants but want to make sure CI works on Github first.

Refactor OpenBSD-like module into OpenBSD

The only OpenBSD-like OS that we support after bitrig's removal is OpenBSD itself.

cc @semarie

[breaking change] gettimeofday 2nd argument incorrect in some targets

The second argument of `gettimeofday` was a `*mut c_void` on all targets,
but that type is incorrect in the following targets, where it should be
a `*mut timezone` instead:

On these other targets it appears that the signature of gettimeofday was incorrect (it takes a time-zone pointer instead of a void pointer):

*linux+gnu: http://man7.org/linux/man-pages/man2/gettimeofday.2.html
*freebsd: https://www.freebsd.org/cgi/man.cgi?query=gettimeofday&apropos=0&sektion=2&manpath=FreeBSD+11.2-stable&arch=default&format=html
*openbsd: https://man.openbsd.org/gettimeofday.2
*android: https://github.com/ricardoquesada/android-ndk/blob/master/usr/include/sys/time.h
*dragonfly: https://www.dragonflybsd.org/cgi/web-man?command=gettimeofday&section=2

This commit corrects the type on these targets, which is a breaking change. Due
to how this API is commonly used (e.g. passing `ptr::null_mut` to the second
argument), breakage should be minimal or non-existent (AFAICT only `time`, `libstd`, and `parking_lot` use this API, and they all should compile after this change). Users wanting to support both versions can just write `ptr as *mut _` instead.

Closes #1338.

---

On these targets, the signature of `gettimeofday` was correct (the second argument is a `void*`):

* macosx: https://developer.apple.com/library/archive/documentation/System/Conceptual/ManPages_iPhoneOS/man2/gettimeofday.2.html
* linux+musl: http://git.musl-libc.org/cgit/musl/tree/include/sys/time.h#n11
* linux+newlib: https://chromium.googlesource.com/native_client/nacl-newlib/+/99fc6c167467b41466ec90e8260e9c49cbe3d13c/newlib/libc/include/sys/time.h#74
* netbsd: http://netbsd.gw.com/cgi-bin/man-cgi?gettimeofday+2.i386+NetBSD-8.0
* newlib: https://github.com/devkitPro/newlib/blob/devkitA64/newlib/libc/include/sys/time.h#L370
* solaris/illumos: https://illumos.org/man/3C/gettimeofday
* emscripten: https://chromium.googlesource.com/external/github.com/kripken/emscripten/+/1.35.20/system/include/libc/sys/time.h#11

cc @alexcrichton

The second argument of `gettimeofday` was a `*mut c_void` on all targets,
but that type is incorrect in the following targets, where it should be
a `*mut timezone` instead:

On these other targets it appears that the signature of gettimeofday was incorrect (it takes a time-zone pointer instead of a void pointer):

linux+gnu: http://man7.org/linux/man-pages/man2/gettimeofday.2.html
freebsd: https://www.freebsd.org/cgi/man.cgi?query=gettimeofday&apropos=0&sektion=2&manpath=FreeBSD+11.2-stable&arch=default&format=html
openbsd: https://man.openbsd.org/gettimeofday.2
android: https://github.com/ricardoquesada/android-ndk/blob/master/usr/include/sys/time.h
dragonfly: https://www.dragonflybsd.org/cgi/web-man?command=gettimeofday&section=2

This commit corrects the type on these targets, which is a breaking change. Due
to how this API is commonly used (e.g. passing `ptr::null_mut` to the second
argument), breakage should be minimal. Users wanting to support both versions
can just write `ptr as *mut _` instead.

Closes #1338.