TP

This release bumps the llama.cpp version to 0.2.0 and ggml to 0.21.0, incorporating numerous backend improvements, bug fixes, and new model support across various hardware accelerators like SYCL, OpenCL, Metal, and Vulkan.

This release focuses on stability and correctness within Tensor Parallelism, fixing issues related to 0-sized tensor slices and implementing an AllReduce fallback. Several platform-specific fixes were also applied, including updates to CUDA device handling and context size limits.