Mark SubArrays as invalid outputs for IDAIntegrator(out, t)

[staticfloat]

Let's imagine we have a matrix that we are filling out, column by column with integrator(t), but we realize we want to use the non-allocating version. We try to use integrator(out[:, idx], t), and then realize that of course we need to make that a view because otherwise slicing just creates a copy. Imagine our surprise when the view doesn't work either. Indeed:

julia> out = zeros(3)
       outv = @view(out[:])
       nv = convert(Sundials.NVector, outv)
       nv[1] = 1.0
       out
3-element Vector{Float64}:
 0.0
 0.0
 0.0

The issue, as it turns out, is due to the convert(NVector, x) overload automatically making a copy of non-plain Vector types. I understand the reasoning behind this, as the backing memory store needs to be contiguous memory. It may be possible to write a correct dispatch for SubArrays that are contiguous vectors, but I'm not sure how to do that, and so have opted instead to just make this an error rather than silently return untouched memory.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (81fb59a) 81.80% compared to head (e71b24d) 81.33%.
Report is 8 commits behind head on master.

❗ Current head e71b24d differs from pull request most recent head 23d7738. Consider uploading reports for the commit 23d7738 to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #434      +/-   ##
==========================================
- Coverage   81.80%   81.33%   -0.48%     
==========================================
  Files          11       11              
  Lines        1440     1441       +1     
==========================================
- Hits         1178     1172       -6     
- Misses        262      269       +7     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.