Add models with NMDA dynamics according to Wang (2002)

[janskaar]

Update: Removed draft.

Adds iaf_wang_2002, a model with a conductance based exponential synapse with extra dynamics, from [1]. The model is simplified to allow efficient computation, by taking $t_\mathrm{AMPA,rise} \to 0$ and modelling it as a delta function. Since $t_\mathrm{AMPA,rise} \ll t_\mathrm{AMPA,decay}$, this should not make too much of a difference.

[1] Wang, X. J. (2002). Probabilistic decision making by slow reverberation in cortical circuits. Neuron, 36(5), 955-968. DOI: https://doi.org/10.1016/S0896-6273(02)01092-9

[heplesser]

@janskaar Several formatting test fail. Could you get them fixed?

[janskaar]

@janskaar Several formatting test fail. Could you get them fixed?

All the checks pass now. I have implemented all the changes we discussed, apart from one. I still have both num_ports and state_vec_size variables, as I think the code is a lot easier to understand with separate variables instead of having to do more arithmetic.

I also added the notebook with a comparison between exact/approximate model and the equations in model_details.

[heplesser]

@janskaar Thanks for the updates, but on the CI tests fail now when GSL is available, i.e., the model included. Could you check what is going on? I'd also suggest to merge master since that removes the touchy mini_brunel_ps MPI test.

[jessica-mitchell]

@janskaar The example still needs to be added to the examples/index file. I created one but I cannot find your username is the list for creating PRs

[janskaar]

@janskaar The example still needs to be added to the examples/index file. I created one but I cannot find your username is the list for creating PRs

Perfect, thanks, I pulled it directly from your fork now.