ERROR: error triggered for injection point gin-leave-leaf-split-incomplete

Hi,

I noticed 3 regression test failures like $SUBJECT in cfbot runs for
unrelated patches that probably shouldn't affect GIN, so I guess this
is probably a problem in master.  All three happened on FreeBSD, but I
doubt that's relevant, it's just that the FreeBSD CI task was randomly
selected to also run the "regress-running" test suite, which is a sort
of Meson equivalent of make installcheck (eg running tests against a
cluster that was already running).  As for why it might require
regress-running or have started only recently, it could be due to
yesterday's boost in the number of CPUs for FreeBSD CI, changing the
timing.  (?)

http://cfbot.cputube.org/highlights/regress-7.html