ENH: Use 4+ empty edge planes to detect skull-stripped images

[effigies]

This proposes a relaxed heuristic for detecting skull-stripped images, detecting each empty edge plane, rather than all edge planes at once. If four planes are empty, we consider it skull-stripped.

This accounts for acquisitions where the brain is cut off on one or two edges.

An alternative we could look at is the actual edges, rather than edge planes. These should be pretty reliably 0 if the image is skullstripped.

Ref: #232 (comment)

[codecov]

Codecov Report

Merging #233 (e220db5) into maint/0.7.x (904c027) will decrease coverage by 0.74%.
The diff coverage is 46.15%.

@@               Coverage Diff               @@
##           maint/0.7.x     #233      +/-   ##
===============================================
- Coverage        71.05%   70.31%   -0.75%     
===============================================
  Files               16       18       +2     
  Lines             1071     1088      +17     
  Branches           155      159       +4     
===============================================
+ Hits               761      765       +4     
- Misses             258      270      +12     
- Partials            52       53       +1     

Flag	Coverage Δ
ds005	59.77% <26.92%> (-1.05%)	⬇️
ds054	49.72% <26.92%> (-0.89%)	⬇️

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

Impacted Files	Coverage Δ
smriprep/cli/check_skullstrip.py	27.77% <27.77%> (ø)
smriprep/workflows/anatomical.py	67.12% <50.00%> (-1.51%)	⬇️
smriprep/utils/images.py	100.00% <100.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 904c027...e220db5. Read the comment docs.

[effigies]

False postives:

• https://openneuro.org/datasets/ds000005/file-display/sub-06:anat:sub-06_T1w.nii.gz
• https://openneuro.org/datasets/ds001734/versions/1.0.5/file-display/sub-046:anat:sub-046_T1w.nii.gz (and all of ds001734)
• https://openneuro.org/datasets/ds001486/file-display/sub-069:ses-T2:anat:sub-069_ses-T2_T1w.nii.gz (and a couple others)
• https://openneuro.org/datasets/ds000208/file-display/sub-34:anat:sub-34_T1w.nii.gz (and several others)
• https://openneuro.org/datasets/ds001608/file-display/sub-09:anat:sub-09_T1w.nii.gz