optimal electrode placement based on ROI

Can we include a measure in the user interface that allows to you predict the activity in one region given the activity in another region (i.e. if we want to know activity in the amygdala, can we reconstruct that activity using electrodes in the contralateral occipital cortex, and to what degree of certainty ... to inform potential electrode placement by regions of desired reconstruction.

No files, tests, or entry points are named. Start by locating the user-interface code and the Gaussian process prediction entry point, then determine how regional activity and electrode locations are represented. Done means the interface can estimate activity in one region from another, report certainty, and support electrode-placement decisions.

Written by the indexing model from the issue text.