Bugfix/234

[SeppeStaelens]

This PR fixes the evolution of the gauge variable B in the IntegratedMovingPuncture gauge by adding a new function to both gauge classes that adds matter terms to the evolution equations of the gauge variables.

[KAClough]

This looks great and it compiles and runs fine. Sorry to ask for more, but could we change the Kerr example to use the Integrated puncture gauge, just to give an example of how it works to use a different gauge and have a test that this now runs ok? It is actually the better gauge for Kerr because it is not initially conformally flat, which is where you want the integrated version.

[SeppeStaelens]

Hi Katy, sure I can adapt the Kerr example. Do you mean also keep the change to the IMP gauge in the main repo? And with a test, do you mean just doing a run, checking the results, or actually implementing a test in the continuous integration (i.e. the checks GitHub runs)?

[KAClough]

Yes, sorry! I meant let's update the Kerr example in the main repo so that it uses the IMP gauge. At the moment both vacuum examples use MP which is fine but it is nice to demonstrate that both gauges can work. In terms of a test, just seeing that it runs sensibly, the gauge stabilises and doesn't blow up should be good enough! I guess we also need to set B^I equal to the initial \barGamma^i. It would be nice to have a matter example with IMP too, but we can do that in the future.

[SeppeStaelens]

Ah yes, I see what you mean. I will change the gauge in the Kerr example, and have it run - I will let you know!

[KAClough]

This all looks good and compiles and runs ok.