Automate update of cooling tables when Cloudy is updated

[abensonca]

When Cloudy is updated we need to regenerate the atomic CIE cooling (and chemical state) tables using the new version. This could be automated:

1. Detect a Cloudy update (by looking for changes in the dependencies.yml file;
2. Run the table build task to generate new tables (which will be made in dynamic datasets);
3. Generate a PR for the datasets repo with these new tables (and old tables copied to new files);
4. Generate a new plot of the cooling function;
5. Run the tests.cooling.exe code to determine if we need to update the values of the cooling target values;
6. If we do, generate that patch and push as a new commit to our branch. (Need to be sure here that we don't get into an infinite loop - detect if the update was already made - perhaps by looking for a change in tests.cooling.F90?)
7. Copy the files from dynamic to static so that all other tests will make use of them. (Need to make these tests dependent on the cooling table updater job.)