distributed_scan.py: Jaguar Scan Distributed over Processors

The distributed_scan.py script splits up a 1D or 2D scan job into multiple jobs that can be distributed over multiple processors, and is used automatically by Maestro for such a job. Jobs with more than two scan coordinates must be run sequentially, as a Jaguar job. The script accepts a single input file that defines the scan job. By default, the subjobs generated use a fixed number of canonical orbitals via the ncanorb keyword, this can be turned off with the -no_ncanorb flag. The processors are specified using the -HOST job control option (see Running Jobs From the Command Line). The results are collated into a single output when all jobs have finished.