Reuse code for metadata

[CGDogan]

Summary

There were four copies of OpenSlide metadata code which I need to refactor for BioFormats. Now there's only two but not one because I'm keeping NCISlideUtil untouched.

Testing

Visited http://127.0.0.1:4010/loader/data/one/fileinimagesfolder.tif, http://127.0.0.1:4010/loader/data/many/[%22fileinimagesfolder%22]. Isolated from Omniload.py, ran the modified code:

#import OmniLoad

import dev_utils
# get fields openslide expects
def openslidedata(manifest):
    for img in manifest:
        img['location'] = img.get("path", "") or img.get("location", "") or img.get("filename", "") or img.get("file", "")
        metadata = dev_utils.getMetadata(img['location'], False, True)
        print(metadata)
        for k, v in metadata.items():
          if k not in img:
            img[k] = v
        # required values which are often unused
        img['study'] = img.get('study', "")
        img['specimen'] = img.get('specimen', "")
    return manifest

print(openslidedata([{"path": "/tiff/file.tiff"}]))

Questions

I hope it's OK to leave NCISlideUtil? That accesses some metadata I couldn't find documentation for so I won't be able to find BioFormats equivalents.

[birm]

This looks good. Glad to see this cleaned this up!
I added an optional suggestion, let me know if you'd prefer to merge without it.