Unified References for Integration Tests and Given Examples

[vincent-maillou]

Currently, the reference solution for the given examples/ can either be:

• Ground truth used to generate the data (synthetic dataset)
• R-INLA convergence
• INLA_DIST convergence
  We need per model the reference solution (ground truth) + a solution converged by either R-INLA or INLA_DIST.

The list (to keep updated) of the examples where this needs to be done follows:

• gr
• gs_coreg2_small
• gs_coreg3_small
• gst_coreg2_small
• gst_coreg3_small
• gst_large
• gst_medium
• gst_small
• pr
• pst_small