Thank you!
This should also include a figure in included_decision_trees.rst (I can help make the figure) and an update to "[tedana] What are the differences between the tedana_orig, meica, and minimal decision trees?" in faq.rst. Hopefully we can have some people test & make sure this is useful before making those changes.

I ran tedana on some infant data where the kappa and rho elbow criteria were reported to be causing problems. In my hands, it was possible to get plausible elbow estimates. There were other issues that might be better address through things like adding censoring (#1053). It's also probably that we can improve the decision tree for an infant application, but I didn't see evidence that an elbow-less minimal tree would be the main fix for their challenges.

Unless we find a use case where this addition clearly helps, I'd suggest closing this PR and maybe the related issue for now.