Comparative performance of neurosurgery-specific, peer-reviewed versus general AI chatbots in bilingual board examinations: evaluating accuracy, consistency, and error minimization strategies.