MORABLES: A Benchmark for Assessing Abstract Moral Reasoning in LLMs with Fables