Error summary in output

The current output can be really unhelpful when there are heaps of errors. By default you should get a breakdown of error code counts and errors per file.

The issue does not name files, tests, or an entry point. Start by locating the output-reporting code and existing tests for error display, then determine the expected default format for error-code counts and errors grouped by file. Done means the default output provides both breakdowns when many errors are present.

Written by the indexing model from the issue text.