Red Teaming in LLMs: Enhancing AI Security and Resilience