feat(runner): the summary says how many steps the generator drove

A green llm run carried no evidence of how much the generator actually
drove: the count was inferable only from llm-calls.jsonl outcomes, and
the number the refusal turns on was invisible in the run's own output.

internal/runner/testdata/violation-summary.txt

@@ -1,5 +1,5 @@
 
-run complete: 4 steps
+run complete: 4 steps, 0 driven by the generator
 1 violation record(s):
   step 1: [balanceNonNegative]
 4 action(s) never reached the app: no_action_produced 4