CUDA: add softmax broadcast#14475

am17an merged 4 commits intoggml-org:gg/ggml-batch-soft-max-opsggml-org/llama.cpp:gg/ggml-batch-soft-max-opsfrom

am17an:add_softmax_batch_and_broadcastam17an/llama.cpp:add_softmax_batch_and_broadcastCopy head branch name to clipboard