Function ple_conv 

pub fn ple_conv(
    gpu: &dyn GpuBackend,
    kernel: KernelHandle,
    x: DevicePtr,
    gated: DevicePtr,
    weight: DevicePtr,
    state: DevicePtr,
    out: DevicePtr,
    num_tokens: u32,
    channels: u32,
    k_size: u32,
    dilation: u32,
    stream: u64,
) -> Result<()>

2026-09-25: Depthwise causal conv with kernel k_size and dilation dilation, then out = gated + silu(conv). All buffers but weight (BF16) are FP32. state is [(k_size - 1) * dilation, channels] and is rolled in place, so prefill and decode use the same launch.