Small improvements to mlVARsim()

[bsiepe]

Description

• Fixed small error where the random effects SD wasn't indexed correctly
• Added the possibility to manually set the sparsity of the temporal and contemporaneous matrices (and updated the documentation accordingly)

This closes #22

Example/small test

The function now returns the same SD for all columns of the temporal matrix, as intended:

set.seed(4321)
library(mlVAR)
sim <- mlVARsim(
  nPerson = 100, 
  nNode = 4, 
  nTime = 100, 
  init_beta_SD = c(1, 1)
  )
sim$model$Beta$SD[,,1]

          [,1]      [,2]      [,3]      [,4]
[1,] 0.2287679 0.2287679 0.2287679 0.2287679
[2,] 0.2287679 0.2287679 0.2287679 0.2287679
[3,] 0.2287679 0.2287679 0.2287679 0.2287679
[4,] 0.2287679 0.2287679 0.2287679 0.2287679