Support Evaluation for the TREC 2024 RAG Track: Comparing Human versus LLM Judges