Add ability to call/omit specific steps of the pipeline

[bhilbert4]

In shared_tasks/run_pipeline.py, the code assumes that all steps of calwebb_detector1 should be run on a given input file (with the exception of ramp fitting).

For the dark monitor, we want to skip the dark_current step. Further, there are _dark.fits files in the filesystem. Starting from these, only the jump and ramp-fitting steps would have to be run, which would save time compared to running the entire pipeline on the uncal version of these files.

It would be helpful to tweak the code such that lists of pipeline steps to skip/run could be provided, and the pipeline call would be updated accordingly.

Finally, at the moment the code is using the run() method of the pipeline. We should switch to using the call() method so that the pipeline will use the appropriate parameter reference files for parameters that we don't explicitly specify. This would involve switching the parameters we do specify to use a nested dictionary.

e.g.:
parameters = {'dark_current': {"skip": True}}
p = Detector1Pipeline.call(uncal_file, steps=parameters}

FYI @york-stsci