Function evaluate_regression 

pub fn evaluate_regression<M, Met>(
    model: &mut M,
    metric: &mut Met,
    samples: impl IntoIterator<Item = RegressionSample>,
) -> Result<Option<f64>, RillError>
where
    M: OnlineRegressor,
    Met: Metric<Truth = f64, Prediction = f64>,

Run progressive evaluation on a regression stream.

The model is updated in place. Returns the final metric value.

The evaluation order is strictly predict → metric.update → learn.