Possible incorrectness in VDP example Jacobian ordering / docs

[skyreflectedinmirrors]

CC: @kyleniemeyer @andrewalferman

So Andrew's work has show that the Jacobian ordering in the VDP problem is possibly incorrect (and at the very least the docs are confusing / misleading)

Todo:

• Verify expected Jacobian order from pyJac is what is described in accelerInt docs
• Verify that VDP example uses correct Jacobian order
• Check that "Column major" references in docs correspond to expected VDP order

I think we actually ran into this issue with @stonecp a few months ago?

see the fixed oregonator Jacobian here

[stonecp]

I did have trouble using the Jacobian. The matrix index ordering is confusing as it seems to depend upon the target platform. It only stores K-1 species (using mass conservation) but that is not consistent with the RHS function itself.

[kyleniemeyer]

Bump?

[skyreflectedinmirrors]

@kyleniemeyer So I had updated the pyJac docs previously, and I believe that the documentation branch of accelerInt has fixed this problem, but I believe I was waiting on merging Andrew's Oregonator example before I merged this?

I'll try to take a look at it in the next few days