Remove scikit-learn as a dependency

[tcrasset]

The package is a huge dependency.
Currently, we only use sklearn in one place (in svd.py):

def SVD(
    df: pd.DataFrame, svd_flip: bool = True
) -> Tuple[NDArray[Any], NDArray[Any], NDArray[Any]]:
    """
    ...
    svd_flip: bool
        Whether to use svd_flip on U and V or not.
    ...
    """
    U, s, V = linalg.svd(df, full_matrices=False)
    if svd_flip:
        U, V = sklearn.utils.extmath.svd_flip(U, V)
    return U, s, V

and the function is a couple lines long:

def svd_flip(u, v, u_based_decision=True):
    """Sign correction to ensure deterministic output from SVD.

    Adjusts the columns of u and the rows of v such that the loadings in the
    columns in u that are largest in absolute value are always positive.
    [...]

    """
    if u_based_decision:
        # columns of u, rows of v
        max_abs_cols = np.argmax(np.abs(u), axis=0)
        signs = np.sign(u[max_abs_cols, range(u.shape[1])])
        u *= signs
        v *= signs[:, np.newaxis]
    else:
        # rows of v, columns of u
        max_abs_rows = np.argmax(np.abs(v), axis=1)
        signs = np.sign(v[range(v.shape[0]), max_abs_rows])
        u *= signs
        v *= signs[:, np.newaxis]
    return u, v