[BUG] utils/weighted_metrics.py non arguments

[TonyBagnall]

Describe the bug

some legacy code in weighted_metrics allows None arguments for the weights, then tests weights.ndim, so crashes if you dont pass weights. I would KWF but its used in performance metrics

def _weighted_geometric_mean(y, weights=None, axis=None):
    """Calculate weighted version of geometric mean.

    Parameters
    ----------
    y : np.ndarray
        Values to take the weighted geometric mean of.
    weights: np.ndarray
        Weights for each value in `array`. Must be same shape as `array` or
        of shape `(array.shape[0],)` if axis=0 or `(array.shape[1], ) if axis=1.
    axis : int
        The axis of `y` to apply the weights to.

    Returns
    -------
    geometric_mean : float
        Weighted geometric mean
    """
    if weights.ndim == 1:

Steps/Code to reproduce the bug

from aeon.utils.weighted_metrics import _weighted_geometric_mean
import numpy as np
y = np.array([1.0,2.0, 3.0])
w = _weighted_geometric_mean(y)

Expected results

it should I think just require weights, otherwise whats the point?

Actual results

        geometric_mean : float
            Weighted geometric mean
        """
>       if weights.ndim == 1:
E       AttributeError: 'NoneType' object has no attribute 'ndim'

axis       = None
weights    = None
y          = array([1., 2., 3.])

..\weighted_metrics.py:34: AttributeError

Versions

No response