Regarding an automated way for LR to show all photos with a given face in them at a mouse click:
When the facial recognition routine recognizes “Uncle Chuck” all it does is add that tag in the metadata file for the image. That’s it.
Of course, there may well be 2 or 200 additional metadata tags associated to that same image. In other words, there is no “column in the data table” named “face” that is filled with the entry Uncle Chuck. Rather, the metadata “column” (to use the relational database technical term) can have any number of entries.
Ergo, it is impossible to write code that would magically “know” which tags are face names, and subsequently perform the task you desire.
If Adobe were to change that design, then perhaps some day it could happen. But for now it would simply not be possible.
If you want to reply, then
register here. Registration is free and your account is created instantly, so you can post right away.