Jupyter-scatter

- Fix distributed build by ensuring that `jscatter/bundle.js` is included in the build
- Fix categorical encoding for partial data

> **Warning**: do not use this version! The distributed build is broken. Use `v0.12.6` instead. :pray:

- Ensure that the default point colors respect the background when setting both at the same time during initialization. I.e., in the following scenario, the point color will be set to _white_ by default as the background color was set to _black_:

py
jscatter.plot(data=df, background_color='black')


- Fix an issue when working with views of a pandas DataFrame where not all categorical data is present

- Loosen strictness of `rows` and `cols` of `compose()` to allow having empty cells in the grid

- Respect the dictionary key-value order of categorical encoding maps in the legend. E.g., the following categorical color map legend will read `C`, then `B`, and finally `A`:

py
scatter.legend(True)
scatter.color(map=dict(C='red', B='blue', A='pink'))


- Update third-party JS libraries

- Fix incorrect legend for categorical coloring

- Update `regl-scatterplot` to `v1.6.9`

- Fix the ordering of the legend's value labels for continuous encodings such that high to low values are order top to bottom ([70](https://github.com/flekschas/jupyter-scatter/issues/70))