GST checkpointing misses the final chi-squared iteration

[rileyjmurray]

This code block manages checkpointing for the GateSetTomography class:

pyGSTi/pygsti/protocols/gst.py

Lines 1417 to 1436 in 962216e

	#Now loop through the generator and query the intermediate results:
	#Do all but the last circuit list.
	for i in range(starting_idx, len(bulk_circuit_lists)):
	#then do the final iteration slightly differently since the generator should
	#give three return values.
	if i==len(bulk_circuit_lists)-1:
	mdl_iter, opt_iter, final_objfn = next(gst_iter_generator)
	else:
	mdl_iter, opt_iter = next(gst_iter_generator)
	mdl_lsgst_list.append(mdl_iter)
	optima_list.append(opt_iter)
	if not disable_checkpointing:
	# update the checkpoint along the way:
	checkpoint.mdl_list = mdl_lsgst_list
	checkpoint.last_completed_iter += 1
	checkpoint.last_completed_circuit_list = bulk_circuit_lists[i]
	# write the updated checkpoint to disk:
	if resource_alloc.comm_rank == 0:
	checkpoint.write(f'{checkpoint_path}_iteration_{i}.json')

Note how there's only only one checkpoint for each value of i. This is actually problematic under usual conditions, wherei == len(bulk_circuit_lists)-1 results in two calls to pygsti's optimizer: one call for a chi-squared objective and one call for log-likelihood.

We should change this code so we checkpoint the final GST chi-squared iteration as soon as it's complete (i.e., before the log-likelihood optimization step begins).