feat: Enhance ephemeral review process with reasoning effort handling and retry logic

tests/test_container_runner.py

@@ -72,6 +72,14 @@ def test_build_install_command_includes_configured_reasoning_effort(monkeypatch:
     assert "--reasoning-effort medium" in command
 
 
+def test_build_install_command_can_disable_reasoning_effort_flag() -> None:
+    settings = get_settings()
+
+    command = _build_install_and_run_command(settings, include_reasoning_effort=False)
+
+    assert "--reasoning-effort" not in command
+
+
 def test_chatgpt_mode_requires_existing_auth_json(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> None:
     missing = tmp_path / "missing-auth.json"
     monkeypatch.setenv("CODEX_AUTH_MODE", "chatgpt")
@@ -157,6 +165,59 @@ def test_run_review_ephemeral_api_key_mode_does_not_fallback_to_host(monkeypatch
     assert "API-key auth runner failed" in result["summary"]
 
 
+def test_run_review_ephemeral_retries_without_reasoning_effort_when_unsupported(monkeypatch: pytest.MonkeyPatch) -> None:
+    get_settings.cache_clear()
+    settings = get_settings()
+
+    monkeypatch.setattr(
+        "gitea_codex_bot.workers.container_runner.prepare_review_prompt",
+        lambda *_args, **_kwargs: ("prompt", {"diff": ""}, object()),
+    )
+    monkeypatch.setattr("gitea_codex_bot.workers.container_runner.GiteaClient", lambda _settings: object())
+
+    calls: list[list[str]] = []
+
+    def _fake_run(cmd, *args, **kwargs):
+        calls.append(cmd)
+        if len(calls) == 1:
+            return type(
+                "Completed",
+                (),
+                {
+                    "returncode": 2,
+                    "stdout": "",
+                    "stderr": "error: unexpected argument '--reasoning-effort' found",
+                },
+            )()
+        return type(
+            "Completed",
+            (),
+            {
+                "returncode": 0,
+                "stdout": '{"verdict":"correct","confidence":0.9,"summary":"ok","findings":[]}\n',
+                "stderr": "",
+            },
+        )()
+
+    monkeypatch.setattr("gitea_codex_bot.workers.container_runner.subprocess.run", _fake_run)
+
+    from gitea_codex_bot.types import ParsedCommand
+
+    result, _repo_cfg = run_review_ephemeral(
+        settings,
+        repo="acme/repo",
+        pr_number=1,
+        command=ParsedCommand(name="review", raw="@codex review"),
+    )
+
+    assert result["verdict"] == "correct"
+    assert len(calls) == 2
+    first_shell = calls[0][-1]
+    second_shell = calls[1][-1]
+    assert "--reasoning-effort" in first_shell
+    assert "--reasoning-effort" not in second_shell
+
+
 def test_parse_codex_exec_stdout_from_stream_item_text_json() -> None:
     stdout = '\n'.join(
         [