Function batch_norm 

pub fn batch_norm(
    input: &[f32],
    gamma: &[f32],
    beta: &[f32],
    mean: &[f32],
    var: &[f32],
    epsilon: f32,
) -> Vec<f32>

Batch normalization function (standalone, for backward compatibility)

y = gamma * (x - mean) / sqrt(var + epsilon) + beta