Function fp8_act_scale_to_kmajor 

pub fn fp8_act_scale_to_kmajor(
    gpu: &dyn GpuBackend,
    kernel: KernelHandle,
    a_scale: DevicePtr,
    a_scale_kmajor: DevicePtr,
    m: u32,
    m_pad: u32,
    k: u32,
    stream: u64,
) -> Result<()>

2026-09-25: Launch fp8_act_scale_to_kmajor: copy the quantizer’s row-major [M, K/128] FP32 activation scales into [K/128, M_pad], with zeros in the M..M_pad pad rows. a_scale is only read.

metrale_gpu_runtime::cublaslt::scale_layout holds the same index math and its CPU tests.