Function fp8_gemm_m128_mfast 

pub fn fp8_gemm_m128_mfast(
    gpu: &dyn GpuBackend,
    kernel: KernelHandle,
    input: DevicePtr,
    b_fp8: DevicePtr,
    output: DevicePtr,
    m: u32,
    n: u32,
    k: u32,
    stream: u64,
) -> Result<()>

fp8_gemm_t_m128_mfast: 128-row M tile (2 chunks/CTA), m on the fast axis. Halves the B panel passes relative to fp8_gemm_n128_mfast.