back_obj: Objective function for BACKWASH.
In dcgerard/vicar: Various Ideas for Confounder Adjustment in Regression

This is mostly so that I can use the SQUAREM package.

back_obj(
  par_vec,
  betahat_ols,
  S_diag,
  Amat,
  tau2_seq,
  lambda_seq,
  scale_var = TRUE,
  var_inflate_pen = 0
)

`par_vec`	A huge vector of parameters whose elements are in the following order: pivec, mubeta_matrix, sig2beta_matrix, gamma_mat, muv, Sigma_v, phi, xi
`betahat_ols`	A vector of numerics. The ordinary least squares estimates of the regression coefficients.
`S_diag`	A vector of positive numerics. The standard errors of `betahat_ols`.
`Amat`	The A matrix for the variational EM.
`tau2_seq`	The known grid of prior mixing variances.
`lambda_seq`	A vector of numerics greater than 1. The penalties for the prior mixing proportions.
`scale_var`	A logical. Should we estimate a variance inflation parameter (`TRUE`) or not (`FALSE`)?
`var_inflate_pen`	The penalty to apply on the variance inflation parameter. Defaults to 0, but should be something non-zero when `alpha = 1` and `scale_var = TRUE`.