vulkan: disable coopmat1 flash attention on Nvidia Turing#19290

0cc4m merged 1 commit intomasterggml-org/llama.cpp:masterfrom

0cc4m/vulkan-nvidia-turing-fa-cm1ggml-org/llama.cpp:0cc4m/vulkan-nvidia-turing-fa-cm1Copy head branch name to clipboard