MosaicPipeline fails with UnboundLocalError when provided a ModelLibrary

[braingram]

from romancal.pipeline import MosaicPipeline
from romancal.datamodels import ModelLibrary

lib = ModelLibrary("L3_mosaic_asn.json")  # from regtest files
p = MosaicPipeline()
p(lib)

produces an exception:

File ~/.pyenv/versions/3.12.4/envs/romancal/lib/python3.12/site-packages/stpipe/step.py:585, in Step.__call__(self, *args)
    579 def __call__(self, *args):
    580     warnings.warn(
    581         "Step.__call__ is deprecated. It is equivalent to Step.run "
    582         "and is not recommended.",
    583         UserWarning,
    584     )
--> 585     return self.run(*args)

File ~/.pyenv/versions/3.12.4/envs/romancal/lib/python3.12/site-packages/stpipe/step.py:507, in Step.run(self, *args)
    505     self.prefetch(*args)
    506 try:
--> 507     step_result = self.process(*args)
    508 except TypeError as e:
    509     if "process() takes exactly" in str(e):

File ~/projects/src/romancal/romancal/pipeline/mosaic_pipeline.py:166, in MosaicPipeline.process(self, input)
    163         if input_filename:
    164             result.meta.filename = self.output_file
--> 166 return result

UnboundLocalError: cannot access local variable 'result' where it is not associated with a value

Due to the mosaic pipeline failing to handle all possible returns for filetype. Providing a DataModel causes the same error for the same reason.