Clustergram

- Minor compatibility update for Bokeh plotting.

__API changes__:

- The first argument in ``Clustergram.fit()`` is now ``X`` instead of ``data``.

__Enhancements__:

- API: follow scikit-learn API conventions (59)
- ENH: ensure that x ticks are only integers (58)
- ENH: format time in verbose mode

__Enhancements__:

- ENH: allow weighting by a custom principal component (35)

__Compatibility notes__:

- ``clustergram`` now requires Python 3.8
- RAPIDS.AI implementation has been tested with version 22.12.

__Minor notes__:

- ``examples`` dictionary has been removed. Refer to the notebooks in the documentation.

__Enhancements__:

- ENH: optionally measure BIC during GMM (21)

__Bug fixes__:

- BUG: ``cuML`` non-weighted plot fix (25)

Fix for ``from_data`` method with non-default indices.

__Bugs__:

- BUG: cluster centers empty due to index mismatch (19)

``clustergram`` now supports interactive plotting using a new `.bokeh()` method based on ``BokehJS``. It
can be handy for exploration of larger and more complex clustergrams or those with significant outliers.

__Enhancements__:

- ENH: support interactive ``bokeh`` plots (14)
- ENH: skip ``k=1`` in K-Means implementations (18)

- documentation restructuring