Failing tests which use "target()" now report the highest score
observed for each target alongside the failing example(s), even
without explicitly showing test statistics.
This improves the debugging workflow for tests of accuracy, which
assert that the total imprecision is within some error budget - for
example, "abs(a - b) < 0.5". Previously, shrinking to a minimal
failing example could often make errors seem smaller or more subtle
than they really are (see the threshold problem, and issue #2180).
The canonical version of these notes (with links) is on readthedocs.