Function celu_forward 

pub fn celu_forward<T: Triton, D: Float, const BLOCK_SIZE: i32>(
    x_ptr: T::Pointer<D>,
    y_ptr: T::Pointer<D>,
    n_elements: i32,
    alpha: f32,
)
where
    T::I32Tensor: Tensor<i32, 1> + Comparison<i32, BoolTensor = T::BoolTensor>,
    T::Pointer<D>: AddOffsets<i32, 1, T::I32Tensor, Output = T::Tensor<T::Pointer<D>>>,

Forward: y = max(0, x) + min(0, alpha*(exp(x/alpha) - 1))