Evaluate windowing from dataset when bitsstored = 1 (#1432)

[gofal]

Fixes #1432

Checklist

• The pull request branch is in sync with latest commit on the fo-dicom/development branch
• I have updated API documentation
• I have included unit tests
• I have updated the change log
• I am listed in the CONTRIBUTORS file

Changes proposed in this pull request:

• when bitsstored is 1 and no windowing data is stored, then fo-dicom evaluates the windowing-parameters based on min/max pixel values. Min is 0 and Max is 1, so windowcenter is 0.5 and windowwidth is 1. But with that formula, everything is evaluted to white.
• In case bitsstored is 1, then the pixeldata has not to be evaluted, the min and max are always 0 and 1. With the values of windowcenter 1 and windowwith 1 the images are rendered correctly

[codecov]

Codecov Report

Attention: Patch coverage is 63.63636% with 8 lines in your changes are missing coverage. Please review.

Project coverage is 76.41%. Comparing base (b45ffe1) to head (ca249f8).
Report is 10 commits behind head on development.

❗ Current head ca249f8 differs from pull request most recent head 632032d. Consider uploading reports for the commit 632032d to get more accurate results

Files	Patch %	Lines
FO-DICOM.Core/Imaging/GrayscaleRenderOptions.cs	63.63%	5 Missing and 3 partials ⚠️

Additional details and impacted files

@@               Coverage Diff               @@
##           development    #1751      +/-   ##
===============================================
- Coverage        76.42%   76.41%   -0.02%     
===============================================
  Files              275      275              
  Lines            25410    25405       -5     
  Branches          3043     3044       +1     
===============================================
- Hits             19419    19412       -7     
- Misses            5062     5064       +2     
  Partials           929      929              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[amoerie]

Any chance we can add the sample DICOM file to the test data and add a unit test that demonstrates the GrayScaleRenderOptions.FromDataset returns the correct result?

[gofal]

I will try to find another sample image that is better suited. This 7MB file has a lot of frames and only from frame 100 above there is any pixel data other than black. If I do not find another sample, then I will add the file provided in the issue.

[amoerie]

If it's too much trouble, don't worry about it. I was just thinking this would be an easy test to add, to ensure we don't break this scenario in the future.

Or alternatively, I suppose we can always just construct a new DicomDataset in memory that contains the minimal amount of needed tags for the unit test.