Move CIVET off the INL network

[lindsayad]

This may be impossible for some NQA-1 reason (hard for me to imagine) or INL security concerns (far more likely) but it's pretty incredible how some idiot playing with our network configurations at the lab (port 22!) can single-handedly torpedo development of not only MOOSE but others who use our testing system as well, e.g. libMesh. (I don't actually know that this is anyone's fault, but I have to conjure up someone to direct my anger at.) This has altered my entire development workflow, e.g. in the last few days I've just pushed up PRs and then just not checked them at all because I know nothing will work.

I'm pretty frustrated. Maybe I wouldn't be quite so frustrated if I was on-site and could actually talk to someone who might know what's going on. But also MOOSE is a world-wide project and an outside developer has to be pretty bewildered at what's going on with our test suite right now.

@idaholab/moose-team

[aeslaughter]

We are working on a solution, including digging up old emails. If they don't have a fix soon we will be escalating this to management.

[permcody]

I rattled some cages this morning. This should be fixed soon if it isn't already.

[brianmoose]

Good to see nothing has changed! 😄

[lindsayad]

Haha awesome to have a @brianmoose drop-in! 😄

[fdkong]

Welcome back :)