Function gdn_decode_f32_strided_norm 

pub fn gdn_decode_f32_strided_norm(
    gpu: &dyn GpuBackend,
    kernel: KernelHandle,
    h_state: DevicePtr,
    query: DevicePtr,
    key: DevicePtr,
    value: DevicePtr,
    gate: DevicePtr,
    beta: DevicePtr,
    z_gate: DevicePtr,
    norm_weight: DevicePtr,
    output: DevicePtr,
    batch_size: u32,
    num_k_heads: u32,
    num_v_heads: u32,
    k_dim: u32,
    v_dim: u32,
    qk_stride: u32,
    v_stride: u32,
    gb_stride: u32,
    z_stride: u32,
    out_stride: u32,
    eps: f32,
    stream: u64,
) -> Result<()>

Strided FP32 GDN decode fused with gated RMS norm.

Same recurrent update as gdn_decode_f32_strided, but writes BF16 post-gated-norm output directly and skips the intermediate FP32 output buffer plus per-token gated-rms-norm launches.