Up to version 1.22.4, groff_mdoc(7) only considered the first word
when comparing section headers. For example, ".Sh SEE ELSEWHERE"
and ".Sh SEE Em ALSO" were considered instances of a SEE ALSO
section. In groff-current, exact matches with no sub-macros are
required. Adjust mandoc behaviour.

While here, also fix a very minor mandoc bug, even though no
detrimental effect of the bug on formatting is known. While using
sub-macros in the .Sh HEAD is bad style, the parsers accept it, so
setting the section attribute on the HEAD needs to act recursively.

Messages of the -Wbase level now print STYLE:. Since this
causes horrible churn anyway, profit of the opportunity to stop
excessive testing, such that this is hopefully the last instance
of such churn. Consistently use OpenBSD RCS tags, blank .Os,
blank fourth .TH argument, and Mdocdate like everywhere else.
Use -Ios=OpenBSD for platform-independent predictable output.

inevitable churn caused by the section title change

Using macros in .Sh header lines, or having .Sm off or .Bk -words open
while processing .Sh, is not at all recommended, but it's not strictly
a syntax violation either, and in any case, mandoc must not die in an
assertion. I broke this in rev. 1.124.

Crash found while trying to read the (rather broken) original 4.3BSD-Reno
od(1) manual page.

Complete overhaul of the mandoc(1) test suite.
* Test the recently introduced -Tman output mode, too.
* Specify date and OS arguments in all tests.
* Remove the kludge of sed(1)ing away the page footer lines.
* New make(1) variables SKIP_TMAN, SKIP_GROFF.
* Drop obsolete, clumsy make(1) variable GROFF_TARGETS.
* Delete obsolete mdoc/Bl/E*.sh error reporting tests.
* Silence char/N/basic.
* New targets:
- ascii, ascii-clean - to run -Tascii tests only
- tman, tman-clean - to run the new -Tman tests only
- obj-clean - maintainer only, needed before groff-clean and groff
- groff-clean - maintainer only, affects checked-in files

Now that we don't have groff in base any longer, check in the desired output,
such that the regression suite does not depend on ports.

Messages of the -Wbase level now print STYLE:. Since this
causes horrible churn anyway, profit of the opportunity to stop
excessive testing, such that this is hopefully the last instance
of such churn. Consistently use OpenBSD RCS tags, blank .Os,
blank fourth .TH argument, and Mdocdate like everywhere else.
Use -Ios=OpenBSD for platform-independent predictable output.

inevitable churn caused by the section title change

Using macros in .Sh header lines, or having .Sm off or .Bk -words open
while processing .Sh, is not at all recommended, but it's not strictly
a syntax violation either, and in any case, mandoc must not die in an
assertion. I broke this in rev. 1.124.

Crash found while trying to read the (rather broken) original 4.3BSD-Reno
od(1) manual page.

Complete overhaul of the mandoc(1) test suite.
* Test the recently introduced -Tman output mode, too.
* Specify date and OS arguments in all tests.
* Remove the kludge of sed(1)ing away the page footer lines.
* New make(1) variables SKIP_TMAN, SKIP_GROFF.
* Drop obsolete, clumsy make(1) variable GROFF_TARGETS.
* Delete obsolete mdoc/Bl/E*.sh error reporting tests.
* Silence char/N/basic.
* New targets:
- ascii, ascii-clean - to run -Tascii tests only
- tman, tman-clean - to run the new -Tman tests only
- obj-clean - maintainer only, needed before groff-clean and groff
- groff-clean - maintainer only, affects checked-in files

Now that we don't have groff in base any longer, check in the desired output,
such that the regression suite does not depend on ports.