Cosmological time integration

This is an absolute monster merge request.

It adds support for cosmological time integration. It has barely been tested for accuracy thus far. But I'd like to merge it in so we can move on with the accuracy tests. In principle, nothing is broken in the normal mode of operations and we should be able to run anything we currently do without any problems.

If this is confirmed and merged in, then I will update the examples we have to actually use the new module.

added 1 commit

• 1d2aeda4 - Also include the changes to the symmetry test.

Compare with previous version

And obviously I had to forget a file at the first attempt...

added 1 commit

• 835c678a - Also initialise the cosmology structure in the test125.

Compare with previous version

Oh and I should mention that the changes made only affect the hydro calculation when disabling vectorization.

added 1 commit

• 32077428 - Move cosmology struct declaration into global scope

Compare with previous version

added 4 commits

• 73e4beb4 - Print the min and max active time-bin at every time-step
• 2df4bac8 - Print the engine policy to the snapshots.
• 8532b39d - Print the cosmological model to the HDF5 snapshots.
• 980a2ec5 - Merge branch 'cosmology' of gitlab.cosma.dur.ac.uk:swift/swiftsim into cosmology

Compare with previous version

Sorry, I have just pushed more changes. These are all of cosmetic nature.

added 1 commit

• ede2d290 - Report GSL success correctly

Compare with previous version

No worries, just pushed another small fix to reporting the state of GSL. It might be better to use a --with-gsl option as that is helpful when GSL isn't installed in standard place, but no need to do that for the moment.

added 1 commit

• 90463b78 - Fix syntax typo

Compare with previous version

Think I'm happy to accept this now. I'll leave the branch for you to deal with or not.

merged

mentioned in commit 30c5a76c

mentioned in issue #375 (closed)