“Calcic amphibole equilibria and a new amphibole-plagioclase geothermometer”—reply to the comment of Poli and Schmidt