Anyone remember why META-INF/orm.xml is handled specially in scanning? I am having some trouble locally with scanning matching multiple sources named META-INF/orm.xml and am trying to understand the reason we do what we do here.