The "loop logger" was originally added to try to make the log output a little more user friendly, however, having been using this on a large stacker.yaml for a while now, I feel like it hasn't really achieved that goal and hasn't scaled well.

There's a few problems with the loop logger:

1. Since the status is ordered with dependencies first, on a large stack config you won't actually see any useful output until stacker reaches leaf nodes. So you're basically stuck seeing a bunch of "pending" messages for minutes on end.
2. Ansi escape sequences don't allow for moving up past the top of the terminal. This means that after a stacker build on a large stacker config, if you scroll up you'll see a lot of cut off input.
3. For me at least, it produces a lot of jitter when stacker writes out new statuses, since it has to write everything to the terminal again.
4. When used with -i, you get a ton of useless log messages, since the entire list of steps is output at every status change.

I wonder if just a simple log message per status update would be better, and more scalable, in the long run. As an example, this is what I'm thinking: