Strategic Intelligence in Large Language Models: Evidence from evolutionary Game Theory