feat(autofix): Push PR iteration changes and sync completion webhook

When a PR iteration run finishes, the agent's new file patches still need
to be pushed to the existing PR. Trigger that push from the completion
hook so the PR is updated, and only emit the iteration completion webhook
on the synced pass so the payload reflects the updated PR and we don't
double-fire.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

Author: joseph-sentry

src/sentry/seer/autofix/on_completion_hook.py

@@ -186,6 +186,15 @@ def _send_step_webhook(
         elif current_step == AutofixStep.PR_ITERATION:
             # PR iteration only runs against an existing PR, so there must be pr states.
             assert state.repo_pr_states, "PR iteration completed without any repo PR states"
+            # The iteration agent produces new file patches that still need to be
+            # pushed to the existing PR. The first hook fire (right after the agent
+            # finishes) has unsynced changes; _maybe_continue_pipeline triggers the
+            # push, which re-fires this hook once the PR is updated. Only emit the
+            # completion webhook on that synced pass so the payload reflects the
+            # updated PR and we don't double-fire.
+            _, is_synced = state.has_code_changes()
+            if not is_synced:
+                return
             webhook_action_type = SeerActionType.ITERATION_COMPLETED
             iteration_index = get_latest_iteration_index(state)
             webhook_payload["pull_requests"] = cls._format_pull_requests_payload(state)
@@ -410,6 +419,14 @@ def _maybe_continue_pipeline(
                     },
                 )
 
+        # PR iteration runs against an existing PR rather than the automated
+        # pipeline. Once the agent finishes iterating, push the new changes to
+        # update that PR. _push_changes is a no-op once the repos are synced, so
+        # the hook re-fire after the push doesn't loop.
+        if current_step == AutofixStep.PR_ITERATION:
+            cls._push_changes(group, run_id, state)
+            return
+
         if stopping_point is None or reached_stopping_point:
             # We've reached the stopping point
             return