fix(plane): prepare workspaces only after the start request

On the live cluster the case clone ran at boot: the golden boot had no egress
policy, and the real actor was still being restored, so both TLS handshakes to
GitHub were cut. After the start the actor is RUNNING under its policy (the
delivery the same minute went through). Adds the one-case check Run (a subset
of validation-46) and re-pins the task manifests to the current runner digest.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01EmM1j46qgKMh3zqD1QMwEo

tests/test_plane_runner.py

@@ -758,3 +758,16 @@ def test_the_golden_boot_never_runs_the_command(env: dict[str, str], receiver: R
             proc.wait()
     assert len(counter.read_text().splitlines()) == 1
     assert files_containing(tmp_path, SECRET) == [], "no log, workspace, output or state file holds the credential"
+
+
+def test_workspaces_are_prepared_only_after_the_start_request(env: dict[str, str], receiver: Receiver, tmp_path: Path) -> None:
+    """Substrate's golden boot never gets a start, so it must touch no workspace (and no network): a clone at boot
+    failed on the live cluster while the actor was still being restored (2026-09-29)."""
+    thread, environ, codes = main_in_thread(tmp_path, OUSAST_ARTIFACT_URL=receiver.url)
+    time.sleep(0.5)
+    written = [p for p in (tmp_path / "ws").rglob("*") if p.is_file()] if (tmp_path / "ws").exists() else []
+    assert written == [], "nothing is materialised before the start request"
+    assert post_start(environ["AX_RUNNER_BOUND_PORT"], {"run": "run-1", "task": "stub"})[0] == 202
+    thread.join(timeout=30)
+    assert codes == [0]
+    assert [p for p in (tmp_path / "ws").rglob("*") if p.is_file()], "the start materialises the workspace"