Measure energy in perf lab scripts

[holly-cummins]

Now that we have the automation working (yay!), we can go to the next step and start measuring energy as well.

I'm assuming we'll replicate what John O'Hara did and access RAPL directly. I don't know if we have access to his actual scripts for that part anymore. Alternatively, for more portability, we could use the approach from #38. That has the advantage of not being restricted to intel hardware, but the disadvantage that it's introducing another moving part into the process, and thus another bit of methodology that needs validation.