Add configuration for how job handles multiple datasets

[TickleThePanda]

See icatproject/topcat_ijp_plugin#22

A configuration attribute for jobs should be added which specifies whether the job type, when given multiple datasets

• Runs one job per dataset
• Runs one jobs on all datasets
• Can do either