CAA generated with either a multinomial or logistic normal observation model from retained catch-at-age
array

`simCAA(nsim, yrs, n_age, Cret, CAA_ESS, CAA_nsamp)`

## Arguments

- nsim
Number of simulations

- yrs
Number of years

- n_age
Number of age classes

- Cret
Retained Catch at age in numbers - array(sim, years, maxage+1)

- CAA_ESS
CAA effective sample size. If greater than 1, then this is the multinomial distribution sample size.
If less than 1, this is the coefficient of variation for the logistic normal distribution (see details).

- CAA_nsamp
CAA sample size

## Details

The logistic normal generates the catch-at-age sample by first sampling once from a multivariate normal distribution
with the mean vector equal to the logarithm of the proportions-at-age and the diagonal of the covariance matrix is the square of the
product of the CV and the log proportions (all off-diagonals are zero). The sampled vector is then converted to proportions
with the softmax function and expanded to numbers (CAA_nsamp). This method allows for simulating fractional values in the
catch-at-age matrix.