ENH: Do not raise error in all instances of b-vecs/vals inconsistencies

[oesteban]

We raised a ValueError whenever the number of b-vals and all-zero b-vecs
didn't match. Few lines later, we would overwrite nonzero b-vecs to
match the b-vals, but that step was never reached. This PR allows the
function to just throw a warning instead.

Resolves: #82

[pull-assistant]

Best reviewed: commit by commit

---

Optimal code review plan

     ENH: Do not raise error in all instances of b-vecs/vals inconsistencie...

Powered by Pull Assistant. Last update d1b93aa ... d1b93aa. Read the comment docs.

[oesteban]

This is working locally. Although I'm not super-satisfied with the logger I've assigned this warning, I think it will do for now.

[codecov]

Codecov Report

Merging #100 into master will decrease coverage by 0.82%.
The diff coverage is 60.00%.

@@            Coverage Diff             @@
##           master     #100      +/-   ##
==========================================
- Coverage   51.64%   50.81%   -0.83%     
==========================================
  Files          21       21              
  Lines        1218     1222       +4     
  Branches      161      162       +1     
==========================================
- Hits          629      621       -8     
- Misses        578      586       +8     
- Partials       11       15       +4     

Impacted Files	Coverage Δ
dmriprep/utils/vectors.py	90.24% <60.00%> (-4.14%)	⬇️
dmriprep/workflows/dwi/base.py	30.00% <0.00%> (-16.67%)	⬇️

---

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 2802c50...d1b93aa. Read the comment docs.

[oesteban]

Okay, at the risk of merging some suboptimal code I'll go ahead so that we have more stuff to revise at tomorrow's meeting (also I'm feeling some urgency of fixing master for this new dataset so that I can continue with #97).