Hub unstable detection of running user pods - deletes running user pods

[consideRatio]

There is a step in the startup of the Hub where the hub checks for running pods. It relies on the same mechanism as it does after starting a user pod I think. It seems like this behaves quite unreliably for my deployment. A lot of hub restarts leads to a user here and there gets kicked out after failing to be found.

This is not fun to reproduce as I mess with my users, so for now there is only an example of a typical success rather than failure.

Hub pod logs from startup, a remembered user pod was found to be running.

[D 2019-08-20 09:31:01.007 JupyterHub app:1875] Verifying that erik@redacted-domain.com is running at http://10.72.6.16:8888/user/erik@redacted-domain.com/
[D 2019-08-20 09:31:01.014 JupyterHub utils:218] Server at http://10.72.6.16:8888/user/erik@redacted-domain.com/ responded with 302
[D 2019-08-20 09:31:01.014 JupyterHub _version:60] jupyterhub and jupyterhub-singleuser both on version 1.0.0
[I 2019-08-20 09:31:01.014 JupyterHub app:1888] erik@redacted-domain.com still running

[consideRatio]

I think this was solved by #1401 and #1404 :D