A Survey on Data Contamination for Large Language Models