refactor deprecate *LAST constants fitting #3131

[dybucc]

Description

Constants matching the *LAST naming scheme, and whose purpose fits that of a "delimitter" value have been deprecated as part of the deprecation efforts stemming from #3131.

This patch starts deprecation of symbols matching one of the most common suffixes in identifiers for constants whose value fits that of a "limit." The changes here follow from those in #3620, with the recommended fixes by Amanieu (not pinging for now.)

Notes

I couldn't look into the QNX _NTO_HOOK_LAST constant because I could not find any information on the headers in that target.

While looking into the UTX_DB_LASTLOG constant in the DragonFly BSD upstream repo, I realized there was no symbol by that name. There is, though, a very similar constant, going by the identifier of UTX_DB_LASTLOGX (note the trailing X.) It does not seem like that one symbol has been changed recently judging by the git blame output of the containing file. The last modification to the part of the header where the constant is declared was 16 years ago. We may be defining a misspelled constant in the libc crate, though I can open a PR changing it if that's, indeed, the case.

In the Linux source, I was hesitant to deprecate NF_IP_PRI_SELINUX_LAST the same way I did with NF_IP_PRI_LAST because that does not seem to be a trailing enumeration member nor does it necessarily imply that it exists for the purpose of serving as a limit/bound on some other value(s). Further, looking into section "Priority within hook" of [1], I found that this SELINUX-tagged symbol did have some significance, so I've not deprecated it even though no comments were made against its removal in #3620.

I also couldn't get hold of Solaris sources, though I assumed that the _PC_LAST constant served the same purpose as in illumos, where I did get to check the headers.

I'm not sure if we should deprecate the LC_ALL_MASK constant in OpenBSD, as that seems to be used as a simpler way of getting a bitmask with all bits for each of the LC_ constants set.

Sources

• FreeBSD manpage detailing how UTXDB_LASTLOGIN is not the type of constant we would want to deprecate.

  https://man.freebsd.org/cgi/man.cgi?query=getutxent&sektion=3

• DragonFlyBSD sources showing how they define a UTX_DB_LASTLOGX symbol but not a UTX_DB_LASTLOG symbol as we do.

  https://github.com/DragonFlyBSD/DragonFlyBSD/blame/59b78e44f96799c5f46fb3d6d616c18d838a75d0/include/utmpx.h#L89

• Android sources showing how they use the NF_*_PRI_LAST in much the same manner as the Linux sources (further below.)

  https://cs.android.com/search?q=NF_BR_PRI_LAST&ss=android%2Fplatform%2Fsuperproject
  https://cs.android.com/search?q=NF_IP_PRI_LAST&sq=&ss=android%2Fplatform%2Fsuperproject
  https://cs.android.com/search?q=NF_IP6_PRI_LAST&sq=&ss=android%2Fplatform%2Fsuperproject

• Linux sources showing the use of NF_*_PRI_LAST constants as trailing enumeration constants defined in terms of large values.

  https://github.com/torvalds/linux/blob/eb3f4b7426cfd2b79d65b7d37155480b32259a11/include/uapi/linux/netfilter_bridge.h#L42
  https://github.com/torvalds/linux/blob/eb3f4b7426cfd2b79d65b7d37155480b32259a11/include/uapi/linux/netfilter_ipv4.h#L44
  https://github.com/torvalds/linux/blob/eb3f4b7426cfd2b79d65b7d37155480b32259a11/include/uapi/linux/netfilter_ipv6.h#L47

• Illumos sources showing how _PC_LAST should be changed when adding new _PC constants.

  https://github.com/illumos/illumos-gate/blob/7124a0cccb6ebf2bcffdab36f9bdf8fe4c23d77f/usr/src/uts/common/sys/unistd.h#L336

• OpenBSD sources showing how _LC_LAST serves as the "end" value for constants defined prior to it.

  https://github.com/search?q=repo%3Aopenbsd%2Fsrc%20_lc_last&type=code

Checklist

• Relevant tests in libc-test/semver have been updated
• No placeholder or unstable values like *LAST or *MAX are included (see #3131)
• Tested locally (cd libc-test && cargo test --target mytarget); especially relevant for platforms that may not be checked in CI

@rustbot label +stable-nominated

[rustbot]

Some changes occurred in a NetBSD-like module

cc @semarie

Some changes occurred in an OpenBSD module

cc @semarie

Some changes occurred in a solarish module

cc @jclulow, @pfmooney

Some changes occurred in an Android module

cc @maurer

[dybucc]

CI actually passes. There seems to be an issue with a glob import that is not used, but this has not
been changed in the patch (it's not even part of it, for that matter.) For some reason, rebasing
onto main with dependabot updates has ended up with a warning across all of my open PRs due to
that one (now apparently unused) import.

[rustbot]

This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.

[tgross35]

For updating doc comments like the others, @rustbot author

[rustbot]

Reminder, once the PR becomes ready for a review, use @rustbot ready.