All tests should fail for stderr
Even if not validating against a reference, all tests should fail for stderr.
This is essential for exposing the problems in the sanitizer builds, where currently some tests are passing despite issues, e.g. https://lhcb-nightlies.web.cern.ch/nightly/lhcb-sanitizers/984/Moore/x86_64_v2-centos7-gcc11-dbg%2Bubsan/tests#Hlt1Conf.persistency.allen_mdf_write