SummaryBot comments on AnimalHarmBench 2.0: Evaluating LLMs on reasoning about animal welfare