Evaluating Human Trust in LLM-Based Planners: A Preliminary Study