Smarter compression: Tailoring AI with LLM Compressor in OpenShift AI