Function generic_stability 

pub fn generic_stability(
    data: &FdMatrix,
    y: &[f64],
    scalar_covariates: Option<&FdMatrix>,
    ncomp: usize,
    n_boot: usize,
    seed: u64,
    task_type: TaskType,
) -> Result<StabilityAnalysisResult, FdarError>

Generic explanation stability via bootstrap resampling.

Refits the model on bootstrap samples and measures variability of coefficients, β(t), and metric (R² or accuracy).

Note: This only works for regression and logistic models since it requires refitting. For classification models, bootstrap refitting is not yet supported.

Errors

Returns FdarError::InvalidDimension if data has fewer than 4 rows, zero columns, or y.len() != n. Returns FdarError::InvalidParameter if n_boot < 2, ncomp is zero, or task_type is MulticlassClassification. Returns FdarError::ComputationFailed if not enough bootstrap refits succeed.