fix(health): akka check uses the shared SPEC Default policy, not Degraded-only OtOpcUaCompat (lmxopcua#771)

OtOpcUaCompat was a migration shim (1d729fb0) preserving the pre-adoption
two-way mapping; nothing depends on it. A node that is not a live cluster
member now fails /health/ready.

Claude-Session: https://claude.ai/code/session_01GT67AEHd2Acv1YPwGfXEoV

src/Server/ZB.MOM.WW.OtOpcUa.Host/Health/HealthEndpoints.cs

@@ -84,7 +84,11 @@ public static class HealthEndpoints
                 "akka",
                 failureStatus: null,
                 tags: new[] { ZbHealthTags.Ready, ZbHealthTags.Active },
-                args: AkkaClusterStatusPolicy.OtOpcUaCompat)
+                // Shared SPEC Default policy (lmxopcua#771). OtOpcUaCompat was a migration shim that kept the
+                // pre-adoption two-way behaviour (anything not Up = Degraded), which can never fail /health/ready
+                // — Degraded passes. Nothing relies on that: Down/Removed/Joining-failed now answer 503 on the
+                // ready tier, Leaving/Exiting stay Degraded (a draining node keeps serving).
+                args: AkkaClusterStatusPolicy.Default)
             // "Is this node in charge of its mesh?" — Healthy on the one node that owns the active
             // work for its Cluster, Unhealthy (503) on its partner, so Traefik pins the AdminUI to the
             // node actually hosting the cluster singletons.