llama.cpp

ba1ceb34 - vulkan: fix noncontig check for mat_mul_id splitting (#14683)

vulkan: fix noncontig check for mat_mul_id splitting (#14683) * vulkan: fix noncontig check for mat_mul_id splitting Remove supports_op check for > 4096 (splitting fixes this) * vulkan: fix batched matmul dequant for Q*_K