Added DICOM conversion script

[bsekachev]

Motivation and context

CVAT does not support DICOM format natively.
One way to annotate DICOM data is to convert them to regular images first. The patch adds a script to support some kinds of DICOM files to .png

How has this been tested?

Manual testing, converted:

• CHAOS training set CT/MT
• Some multi-frame DICOM files from here

Checklist

• I submit my changes into the develop branch
• I have added description of my changes into CHANGELOG file
• I have updated the documentation accordingly
• I have added tests to cover my changes
• I have linked related issues (read github docs)
• I have increased versions of npm packages if it is necessary (cvat-canvas,
  cvat-core, cvat-data and cvat-ui)

License

• I submit my code changes under the same MIT License that covers the project.
  Feel free to contact the maintainers if that's a concern.
• I have updated the license header for each file (see an example below)

# Copyright (C) 2021 Intel Corporation
#
# SPDX-License-Identifier: MIT

[coveralls]

Coverage remained the same at 74.055% when pulling de5715b on bs/added_dicom_conversion_script into 6988e2b on develop.

[nmanovic]

Better to provide an example for a real dataset (minor comment)