fix(#4130): coerce stale 'max' in get_reasoning_status read path too (Codex catch)

get_reasoning_status() returned the raw config reasoning_effort, so a stale saved
'max' surfaced on boot/status/chip even though streaming coerces it to xhigh. Now
routes the status value through coerce_reasoning_effort_for_model() with the resolved
model/provider/base_url, so all read paths agree. Added a stale-max->xhigh status
regression test.

api/config.py

@@ -2734,7 +2734,16 @@ def get_reasoning_status(
     return {
         # Match CLI default (True if unset in config.yaml)
         "show_reasoning": bool(show_raw) if isinstance(show_raw, bool) else True,
-        "reasoning_effort": str(effort_raw or "").strip().lower(),
+        # Report the COERCED effort (not the raw config value) so boot/status/chip
+        # read paths agree with what streaming actually sends — e.g. a stale
+        # `reasoning_effort: max` surfaces as `xhigh`, not the now-unsupported `max`.
+        # (Codex review of the drop-max alignment.)
+        "reasoning_effort": coerce_reasoning_effort_for_model(
+            str(effort_raw or "").strip().lower(),
+            resolve_model,
+            provider_id=resolve_provider,
+            base_url=resolve_base_url,
+        ),
         "supported_efforts": supported_efforts,
         "supports_reasoning_effort": bool(supported_efforts),
     }

tests/test_reasoning_effort_model_capabilities.py

@@ -140,3 +140,20 @@ def test_get_reasoning_status_for_reasoning_capable_model_has_no_max():
     assert status["supported_efforts"] == ["minimal", "low", "medium", "high", "xhigh"]
     assert status["supports_reasoning_effort"] is True
     assert "max" not in status["supported_efforts"]
+
+
+def test_get_reasoning_status_coerces_stale_max_to_xhigh(monkeypatch):
+    """A previously-saved `agent.reasoning_effort: max` (no longer a valid effort)
+    must be reported as the coerced `xhigh`, not the raw stale `max`, so the
+    boot/status/chip read paths agree with what streaming actually sends."""
+    monkeypatch.setattr(
+        cfg,
+        "_load_yaml_config_file",
+        lambda *a, **k: {"agent": {"reasoning_effort": "max"}},
+    )
+    status = cfg.get_reasoning_status(
+        model_id="gpt-5.5",
+        provider_id="openai-codex",
+    )
+    assert status["reasoning_effort"] == "xhigh"
+    assert status["reasoning_effort"] != "max"