Abstract page for arXiv paper 2608.14566: Position: Evaluations of AI Moral Reasoning Still Miss Half of the Picture