Add checkpointing capability to WRF-Hydro #812
Description
Add checkpoint capability to WRF-Hydro to increase resilience from crashes and other unexpected job stops.
The idea here is to update a checkpoint file with each restart, which is then normally deleted at the end of the run. If a run ends prematurely, the job can be started from the same namelists but will know from the checkpoint how much to "fast-forward" before continuing the run. This makes it much easier to automate re-runs without having to do any manual updates of the namelist.hrldas file.