Clarify difference between ArxivP1D, TestSimulation and P1D_MPGADGET

[andreufont]

There is quite a lot of copy paste in there, and things that could be simplified.

I was tempted to do it now, but we should focus on getting the first emulator paper out, so leaving this for the future.

[andreufont]

Treatment of kp_Mpc should also be homogeneized, although right now we don't have parameters.json files for the test simulations. It is probably fine, since we only need to do this for one simulation, and it is not too slow now.

[andreufont]

Do we need to compute the linear power parameters in TestSimulation? We are not going to use this simulation in the emulator... I guess the reason to do this is to pre-compute the linear power parameters from this cosmology, so that we can know what the "emulator calls" should be when making plots like Fig 3 in the draft.
I guess this is particularly useful for IGM parameters, but for cosmology parameters we could compute these on the fly whenever we need them, it should be a couple of seconds now. This way we wouldn't need o pass kp_Mpc to this object, I found this confusing. You should be able to generate a synthetic dataset before you decide which pivot point you will use.

Again, though, not a priority, so leaving the comment here only for future reference.