vulkan: support arbitrary KV dimension in flash attention (llama/16160)

vulkan: support arbitrary KV dimension in flash attention (llama/16160)

The "Clamp" spec constant is already based on whether KV is a multiple of Bc,
so use that to control whether bounds checking is performed. Add bounds checking
to the scalar and coopmat1 paths. Coopmat2 didn't need any changes (the K/V
tensors are already optionally clamped, nothing else needed to be changed).