llama.cpp b10173 adds Laguna-S-2.1 LLM support

The latest release of llama.cpp (b10173) introduces support for the Laguna-S-2.1 large language model, enhancing capabilities for local AI deployments on consumer hardware like Apple Silicon and RTX GPUs. This update also includes binaries optimized for macOS ARM64 and Intel x64 architectures, ensuring broad compatibility across different platforms. Operators should consider downloading this new version to leverage advanced features of the Laguna-S-2.1 model, which can improve conversational fluency and context understanding.