Also tolerate IndexError while trying to delete a field in mw

[yarikoptic]

Closes #633

Somewhat of a blind fix without looking to deep why we get IndexError instead of KeyError

@neurorepro -- can you check if that would be sufficient?

[neurorepro]

Yes will do it now

[codecov]

Codecov Report

Base: 81.34% // Head: 81.29% // Decreases project coverage by -0.06% ⚠️

Coverage data is based on head (24f4d88) compared to base (e075aca).
Patch coverage: 100.00% of modified lines in pull request are covered.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #634      +/-   ##
==========================================
- Coverage   81.34%   81.29%   -0.06%     
==========================================
  Files          42       42              
  Lines        3828     3828              
==========================================
- Hits         3114     3112       -2     
- Misses        714      716       +2     

Impacted Files	Coverage Δ
heudiconv/dicoms.py	85.77% <100.00%> (ø)
heudiconv/tests/test_bids.py	98.36% <0.00%> (-0.33%)	⬇️
heudiconv/bids.py	89.78% <0.00%> (-0.25%)	⬇️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[yarikoptic]

so, do you think this fix is worth it or not really (given that alone doesn't provide remedy) @neurorepro ?

[neurorepro]

@yarikoptic this fix can be useful if the error raised from nibabel can be reduced to throwing an IndexError. However if solving the nibabel problem means also solving the IndexError then in that best case scenario the fix wouldn't even be necessary, and could even be omitted so that to catch other possible IndexError cases without passing them.