In the case of a user not specifying a dispatch value, fallback/use as default the slurm-dispatch method.
This is essentially causing the workflow to have an expectation it is in-use on a head-node of an HPC.
In addition, change the behavior such that if slurm-dispatch is in use for a given configuration, the existence of the needed commands (sbatch, squeue, etc), is checked ahead of attempted executions of pecan code.
In the case of a user not specifying a dispatch value, fallback/use as default the slurm-dispatch method.
This is essentially causing the workflow to have an expectation it is in-use on a head-node of an HPC.
In addition, change the behavior such that if slurm-dispatch is in use for a given configuration, the existence of the needed commands (
sbatch,squeue, etc), is checked ahead of attempted executions of pecan code.