feat(auditlog): extend ISiteStreamAuditClient with IngestCachedTelemetryAsync (#23 M3)

Add the second site→central RPC seam alongside the existing M2
IngestAuditEventsAsync. The Bundle D proto already lit up
IngestCachedTelemetry (CachedTelemetryBatch / IngestAck) so this commit
just plumbs the client-side abstraction:

* ISiteStreamAuditClient gains IngestCachedTelemetryAsync(batch, ct).
* NoOpSiteStreamAuditClient implements it returning an empty IngestAck
  (same shape as M2 — production gRPC client lands in M6).
* SyncCallEmissionEndToEndTests' DirectActorSiteStreamAuditClient stub
  throws NotSupportedException from the new method so a regression that
  accidentally routes a cached packet through the sync stub fails loudly.
* New NoOpSiteStreamAuditClientTests cover the null-guard + empty-ack
  contract for both batch shapes.

Bundle E task E1.

src/ScadaLink.AuditLog/Site/Telemetry/NoOpSiteStreamAuditClient.cs

@@ -38,4 +38,16 @@ public sealed class NoOpSiteStreamAuditClient : ISiteStreamAuditClient
         // Pending until M6's real client (or a Bundle H test stub) takes over.
         return Task.FromResult(EmptyAck);
     }
+
+    /// <inheritdoc/>
+    public Task<IngestAck> IngestCachedTelemetryAsync(CachedTelemetryBatch batch, CancellationToken ct)
+    {
+        ArgumentNullException.ThrowIfNull(batch);
+        // Empty ack — same rationale as IngestAuditEventsAsync. The M3
+        // CachedCallTelemetryForwarder still writes the audit + tracking rows to
+        // the site SQLite stores authoritatively; central-side state only
+        // materialises once M6's real gRPC client (or a Bundle G test stub) is
+        // wired in.
+        return Task.FromResult(EmptyAck);
+    }
 }