Welcome to the forum @bill!
This is described in this tutorial:
https://docs.qiime2.org/2019.10/tutorials/longitudinal/
and a little more about how to interact with that here:
https://docs.qiime2.org/2019.10/tutorials/sample-classifier/
The short answer is: this contains feature importance scores. You can use qiime metadata tabulate to view and rank features by importance score. This is useful for the maturity-index (tells you which features are most predictive of age), but no need with feature-volatility; the volatility visualization already has an interactive barplot that ranks features by importance score.
Sorry you ran into issues with the donut of death; fixing this is on our radar, see here:
Enjoy!
