Function dense_gemm_bf16_pipelined 

pub fn dense_gemm_bf16_pipelined(
    gpu: &dyn GpuBackend,
    kernel: KernelHandle,
    input: DevicePtr,
    weight: &DenseWeight,
    output: DevicePtr,
    m: u32,
    n: u32,
    k: u32,
    stream: u64,
) -> Result<()>

Pipelined tensor-core BF16 GEMM — drop-in faster dense_gemm (kernel dense_gemm_bf16_pipelined): mma.sync.m16n8k16 + cp.async 2-stage, 128x128 tile. ~40x the scalar dense_gemm on large-M shapes (cosine=1.0, same math). Grid: (ceil(N/128), ceil(M/128), 1) Block: (256, 1, 1)