Function bf16_concat 

pub fn bf16_concat(
    gpu: &dyn GpuBackend,
    kernel: KernelHandle,
    a: DevicePtr,
    b: DevicePtr,
    output: DevicePtr,
    n: u32,
    stream: u64,
) -> Result<()>

BF16 concatenation: out[0..N] = a[0..N], out[N..2N] = b[0..N].

Kernel: bf16_concat(a, b, out, N) Grid: (ceil(N/256), 1, 1) Block: (256, 1, 1)