Automatic tests can succeed with syntax errors

[peter.ludemann]

It seems that there’s a solution, and a rather elegant one at that. Thank-you @jan!

Changes in error handling

As a followup to Automatic tests can succeed with syntax errors and similar topics raised in the past I’ve pushed a number of changes: Add global and thread statistics/2 keys errors and warnings to count errors and warnings. This only counts errors and warnings printed using print_message/2 that are not intercepted using message_hook/3. Add an option --on-error= and --on-warning= and related Prolog flags on_error and on_warning that can all have the values print (default), status (continu…