Allow for use of spinup output in calibration and simulation

[btobers]

Enable spinup output to be used in calibration and simulation. Create a flag or similar method to opt for using spinup output.

[btobers]

Currently, pygem.bin.run.run_spinup stores a model_geometry.nc file with the flowline geometries for each year of model spinup. Need to implement a way to take the flowlines at a year of interest and run calibration/simulations from those - but with debris data added. A little clunky, but one possible option is like so:

# instantiate flowline.FileModel object from model_geometry_dynamic_spinup
fmd_dynamic = flowline.FileModel(gdir.get_filepath("model_geometry", filesuffix=f"_dynamic_spinup_pygem_mb"))
# run FileModel to startyear (it will be initialized at `spinup_start_yr`)
fmd_dynamic.run_until(args.ref_startyear);
# write flowlines
gdir.write_pickle(fmd_dynamic.fls, "model_flowlines", filesuffix=f"_{args.ref_startyear}");
# add debris
debris.debris_binned(gdir, fl_str="model_flowlines", filesuffix=f"_{args.ref_startyear}");
# store flowlines
fls = gdir.read_pickle("model_flowlines", filesuffix=f"_{args.ref_startyear}")