add CAIC and EDF calcs

[James-Thorson-NOAA]

adding function for conditional AIC (CAIC) and effective degrees of freedom (EDF), thanks to new derivation by Nan Zheng for generic calculation using AD+Laplace approximation methods in TMB

[James-Thorson]

@seananderson do you want me to look at this again? As I said before, I'd still like an option for the EDF to be incorporated into the print_one_model output (similar to mgcv::summary),

[seananderson]

I've got as far as cleaning things up and passing all R CMD checks and I just incorporated the function into the main branch so it can be used. Getting the pretty printing working is going to take a chunk of time I haven't found yet, but I'll leave the other issue open until that's done. On top of splitting the smoother, time-varying random effect, and IID random effect groups, there's also the delta model dimension to consider. I'll start with the smoothers, which I already have some code lying around for.

[James-Thorson]

Awesome! Please tell me if you want me to share how I distinguished the groups in demo code for tinyVAST.

Also, I just realized that I had a bug in my various versions, and sorry for catching it now! Specifically it should be q <- sum(indx) instead of q <- length(indx). Do you want me to do a quick PR?

[seananderson]

No problem. I've fixed the q/indx bug in the main branch.

Please tell me if you want me to share how I distinguished the groups in demo code for tinyVAST.

Sure, I'll take any ideas!