ExtractSegmentationExtractor improvements

[weiglszonja]

Connected to catalystneuro/neuroconv#128, raise an error when output_struct_name is not found in the .mat file.

• changed assumed newer version of extract files to "1.0.0"
• transpose segmentation images to have height x width for consistency

[CodyCBakerPhD]

transpose segmentation images to have height x width for consistency

From the tutorial at: https://github.com/schnitzer-lab/EXTRACT-public/blob/master/Learning%20materials/Lecture%20Tutorials/Tutorial%201%20-%20Starting%20code/example_tutorial.m#L38

it would seem the spatial weights already follow ROIExtractors convention

[weiglszonja]

@CodyCBakerPhD leaving here as a reminder
when reading extract data in MATLAB, these images would read as height x width; however in h5py they would read as width x height by default, hence the transpose is necessary in this case to be consistent within extractors.

[codecov]

Codecov Report

Merging #210 (52f2643) into master (f9c3041) will increase coverage by 0.14%.
The diff coverage is 96.66%.

@@            Coverage Diff             @@
##           master     #210      +/-   ##
==========================================
+ Coverage   73.04%   73.19%   +0.14%     
==========================================
  Files          33       33              
  Lines        2282     2294      +12     
==========================================
+ Hits         1667     1679      +12     
  Misses        615      615              

Flag	Coverage Δ
unittests	73.19% <96.66%> (+0.14%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
...schnitzerextractor/extractsegmentationextractor.py	94.77% <96.66%> (+0.44%)	⬆️