n_factors is ignored

[m-clark]

Currently n_factors arg in corr_heat is ignored at the point of checking for problematic input for small K column data (e.g. more factors than data would warrant).

  if (!plot_only) {
    if (rlang::is_null(dimnames(cormat))) 
      stop("Need a correlation matrix with row and column names")
    if (!rlang::is_null(n_factors) && nc < 4) {
      n_factors <- 1
    }
    else {
      n_factors <- floor(sqrt(nc))
    }