feat(scheduler-core): build xceadmin scheduler recording client functions (task 10.3)

Adds record_start()/record_stop()/record_status() to new
src/xceadmin/src/xceadmin/scheduler.py, following push_config()'s exact
pipeline shape (load_schema -> follow_redirects -> decode_reply). Deliberately
diverges from push_config()'s "return failure as data" convention because
this task's own Observable text requires raising a typed error on an error
reply, not just connection/decode failure -- a new SchedAdminError
(xcecommon/errors.py) carries the full reply state (error/recording/file/
event_count/started_at) when SchedAdminReply.ok is false.

test_scheduler.py mirrors test_config_push_mock.py's real mock-TLS-server
technique, covering all three actions, redirect-following (standby-to-
primary), and the error-reply-raises-SchedAdminError path across the
error codes reference table.

188 passed, 3 skipped via the real xceadmin test runner (meson compile
-C xce.build xceadmin).

src/xceadmin/tests/test_errors.py

@@ -21,6 +21,7 @@ from xcecommon.errors import (
     NoPrimaryError,
     ReplyDecodeError,
     ReservedNameError,
+    SchedAdminError,
     SchemaNotFoundError,
     TooManyRedirectsError,
     ValidationError,
@@ -36,6 +37,7 @@ ALL_SUBCLASSES = [
     ReplyDecodeError,
     TooManyRedirectsError,
     NoPrimaryError,
+    SchedAdminError,
 ]
 
 
@@ -171,6 +173,31 @@ def test_no_primary_error_carries_host_and_port() -> None:
     assert caught.port == 9991
 
 
+def test_sched_admin_error_carries_error_and_reply_state() -> None:
+    """Requirement 5.10 (declared here, raised by a later task): a
+    ``SchedAdminReply{ok=false, error=...}`` is surfaced as a typed error
+    carrying the reason plus the reply's recording/file/event/started-at
+    state (e.g. "already recording" still reports the current file)."""
+    with pytest.raises(XceAdminError) as exc_info:
+        raise SchedAdminError(
+            "scheduler admin request failed: already recording",
+            error="already recording",
+            recording=True,
+            file="/var/log/xce/current.replay",
+            event_count=42,
+            started_at=1234567890,
+        )
+
+    caught = exc_info.value
+    assert isinstance(caught, SchedAdminError)
+    assert caught.message == "scheduler admin request failed: already recording"
+    assert caught.error == "already recording"
+    assert caught.recording is True
+    assert caught.file == "/var/log/xce/current.replay"
+    assert caught.event_count == 42
+    assert caught.started_at == 1234567890
+
+
 def test_catching_base_type_recovers_message_for_every_subclass() -> None:
     """The entire point of the shared exception contract (design.md,
     xceadmin.cli Implementation Note): a single `except XceAdminError as e`
@@ -186,6 +213,12 @@ def test_catching_base_type_recovers_message_for_every_subclass() -> None:
         (ReplyDecodeError("msg-f", reason="r"), "msg-f"),
         (TooManyRedirectsError("msg-g", hops=3), "msg-g"),
         (NoPrimaryError("msg-h", host="h", port=1), "msg-h"),
+        (
+            SchedAdminError(
+                "msg-i", error="e", recording=False, file="", event_count=0, started_at=0
+            ),
+            "msg-i",
+        ),
     ]
 
     for exc, expected_message in cases: