Corpus Viewer: Color words by weight in topic

**Is your feature request related to a problem? Please describe.**
It is currently not possible to observe how each individual text corresponds to a topic.

**Describe the solution you'd like**
Corpus Viewer could recognize topic input and use word weight to color words by the corresponding topic. I.e. Topic 1 has weights "apple 0.7, banana 0.3" and Topic 2 has weights "basket 0.8, apple 0.2". So the widget would color all the words "apple" and "banana" with the color of Topic 1, all "basket" with the color of Topic 2.

**Describe alternatives you've considered**
No alternatives really. Perhaps a separate widget if Corpus Viewer cannot be extended in such a way.
Also, we should think about colors vs. highlights vs. colored boxes around words.

**Additional context**
This has already been done elsewhere and it looks like this:

Begin at the Corpus Viewer widget and inspect how topic input and word weights are currently represented. Done means words are visually distinguished by their corresponding topic weights, with the chosen colors, highlights, or boxes defined and behavior matching the example in the issue.

Written by the indexing model from the issue text.