CUDA: fix pointer incrementation in FA#14916

JohannesGaessler merged 1 commit intoggml-org:masterggml-org/llama.cpp:masterfrom

JohannesGaessler:cuda-fix-skip-logicJohannesGaessler/llama.cpp:cuda-fix-skip-logicCopy head branch name to clipboard