Isotope bug fixes and tests

[wwieder]

We need C isotopes to work in CESM3, so I'm going to group these card here for clarity. There's enough here that I wonder if we need to assign this to someone and prioritize the work and will create an issue to bring it up at an upcoming SE meeting.

• C isotope fields can have garbage values starting in ctsm5.1.dev008 #1358
• Use of C13/C14 timeseries for non-transient control cases gives incorrect input when run long enough #592
• Add a system test to catch breakages in carbon isotopes #231
• c13/c14 isotope time-series is awkward for 1850 control (convert to streams file) #182
• Issue warning in build-namelist if you are trying to run Carbon isotopes for present day (non-transient) #181
• Logic for initializing C isotopes and maybe other variables incorrect when using init_interp #67

NOTE: We want to make sure the model aborts when you are trying to run isotopes from a finidat file without it (should be one of the above issues). We will also need to create a set of finidat files when CTSM5.2 comes in.