Function dense_gemm_tc_scaled_acc 

pub fn dense_gemm_tc_scaled_acc(
    gpu: &dyn GpuBackend,
    kernel: KernelHandle,
    input: DevicePtr,
    weight: &DenseWeight,
    output: DevicePtr,
    m: u32,
    n: u32,
    k: u32,
    scale: f32,
    stream: u64,
) -> Result<()>

2026-09-25: output[m, n] += scale * bf16(input[m, k] @ weight[n, k]^T) in one pass: dense_gemm_tc with an accumulating epilogue, used for the LoRA expand+fold (lora_delta.rs). It saves the [m, n] scratch that a GEMM followed by bf16_scaled_add writes and reads back.

Same result as that pair: the kernel rounds each dot product to BF16 before it applies scale and adds, as bf16_scaled_add does with a BF16 scratch.