A Comparative Study of Neurosymbolic AI Approaches to Interpretable Logical Reasoning