Sync tool/lib changes from ruby/ruby

Workaround for jruby/jruby#9026.  JRuby attempts to call the entire
`ENV[“PAGER”]` (or `ENV["RUBY_PAGER"]`) as the program name instead
of splitting it into a program name and options, even if stdout and
stderr are redirected to a non-tty.

…Ruby 2.4 and earlier

Fix SyntaxError: unexpected keyword_rescue, expecting keyword_end in Ruby 2.4 and earlier

`Regexp#match?` was introduced at ruby 2.4.

test-unit depends on power_assert that the recent versions require
ruby 3.1 or later.  While test-unit rescues syntax error at loading
power_assert for old ruby versions, warnings for experimental features
are not suppressed, and `assert_separately` fails because stderr is
not empty, by default.  Since adding `required_ruby_version` to
power_assert causes the installation with old rubygems to fail, we
just ignore warnings for experimental features totally.

ruby/ruby@0ca3eed

Replacing PATH with only the build directory breaks the spawned
ruby.exe on MSYS2 UCRT64, where the Windows DLL loader needs PATH to
resolve libgmp-10.dll linked into the ruby DLL. Prepend the directory
instead so the build ruby is still found first.

ruby/ruby@9272524

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

assert_linear_performance() picks CLOCK_THREAD_CPUTIME_ID when it is
available.  That clock belongs to the native thread that reads it, and an
M:N thread moves between native threads, so the two reads around one
measurement can come from different native threads and the elapsed time
comes out negative:

    Expected 0 to be <= -2.2150775009999997.

Reproduced with RUBY_MN_THREADS=2, 1500 fork/waitpid pairs per run: the
Ruby thread changed native thread 1, 9 and 10 times over three runs, and
the clock went backwards 1, 5 and 6 times, by as much as 0.38s.  The
default, =1 and =-1 never moved the main thread and never went backwards.
CLOCK_PROCESS_CPUTIME_ID stayed monotonic across 15 such migrations.

Prefer process CPU time when the M:N scheduler is on.  This only changes
which clock the assertion uses; it does not relax the assertion.

ruby/ruby carries the same change as 0e06087548.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>