Ever: Mitigating Hallucination in Large Language Models through Real-Time Verification and Rectification