split out pre-processing in coffea into separate step

coffea_split_preprocessing.py

run_instance = processor.Runner(
    metadata_cache={},
    executor=processor.FuturesExecutor(
        workers=6, 
                                      ),
    schema=processor.NanoAODSchema,
    savemetrics=True,
    skipbadfiles=False,
)
filemeta = run_instance.preprocess(to_run, treename="Events")
# optionally filemeta = [f for f in filemeta] to materialize the generator to eg save it 
hists = run_instance.run(filemeta, processor_instance=MyZPeak(), treename="Events")