setup_run_tmb: Set up and run TMB
In njhenry/covidemr: Tools for Mapping COVID Excess Mortality

Generic TMB model run handler. Sets up the ADFun object, applies model speedups and fixes as specified, and optimizes using 'nlminb'. This is meant to be a helper function run by more specific model functions.

setup_run_tmb(
  tmb_data_stack,
  params_list,
  tmb_random,
  tmb_map,
  tmb_outer_maxsteps,
  tmb_inner_maxsteps,
  normalize = FALSE,
  run_symbolic_analysis = FALSE,
  set_limits = FALSE,
  limit_max = 10,
  limit_min = -limit_max,
  parallel_model = FALSE,
  optimization_method = "nlminb",
  model_name = "model",
  verbose = FALSE,
  inner_verbose = FALSE
)

`tmb_data_stack`	List containing all data inputs expected in the TMB CPP file
`params_list`	List containing all parameters expected in the TMB CPP file
`tmb_random`	Character vector containing all random effects that will be optimized in the inner optimizer
`tmb_map`	Named list containing parameters that will be treated in a particular way by the optimizer
`tmb_outer_maxsteps`	Max number of steps taken by the outer optimizer
`tmb_inner_maxsteps`	Max number of steps taken by the inner optimizer in a single outer optimizer step
`normalize`	[boolean, default FALSE] Run TMB's automatic process normalization function? Adds two additional items to the TMB data stack: 'auto_normalize' (1) and 'early_return' (0). Both are used by the `normalize_adfun` function to get the normalizing constant prior to optimization. For more details, see