Resample parameters in outlier detection don't have same limits as in resample

[braingram]

One example (I'm not sure if there are more) is pixfrac which has no min/max in outlier detection: https://github.com/spacetelescope/jwst/blob/619cb3bbf6e6dd122ad807cf73705c5c81c49c06/jwst/outlier_detection/outlier_detection_step.py#L55 unlike in resample: https://github.com/spacetelescope/jwst/blob/619cb3bbf6e6dd122ad807cf73705c5c81c49c06/jwst/resample/resample_step.py#L43 So with the current outlier detection providing a <0 or >1 pixfrac won't produce a ValidationError.

Skimming the 2 steps suggests the following also don't match:

• weight type
• kernel
• fillval