Fix norm bug in curvature computation

[unalmis]

Looks like the matrix norm was being computed instead of the vector norm for normal vector of a rho surface. Affects all curvature quantities. Might be a good idea to review other places we use norm. grep -r -F 'np.linalg.norm'

[codecov]

Codecov Report

Patch coverage: 100.00% and project coverage change: -0.02 ⚠️

Comparison is base (00d255e) 94.16% compared to head (777599f) 94.15%.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #566      +/-   ##
==========================================
- Coverage   94.16%   94.15%   -0.02%     
==========================================
  Files          75       75              
  Lines       16963    16951      -12     
==========================================
- Hits        15973    15960      -13     
- Misses        990      991       +1     

Impacted Files	Coverage Δ
desc/compute/_geometry.py	100.00% <ø> (ø)
desc/compute/_basis_vectors.py	100.00% <100.00%> (ø)
desc/compute/_field.py	100.00% <100.00%> (ø)
desc/plotting.py	87.31% <100.00%> (-0.08%)	⬇️

... and 1 file with indirect coverage changes

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

[f0uriest]

Yeah I had realized recently that n and n_rho were duplicated. I'd prefer to keep the name n_rho to avoid confusion with other normal vectors, other than that looks good.