Text this: Comparative efficacy of ChatGPT 3.5, ChatGPT 4, and other large language models in gynecology and infertility research