fix: reduce reconnect redeploy and metrics load (#476)

* fix: reduce reconnect redeploy and metrics load

Throttle node-online redeploy retries and lower the agent metric cadence so brief reconnect churn no longer fans out into repeated runtime syncs and backend connection pressure.

* docs: add follow-up implementation design notes

Document the planned flow upload batching work and the local remote-address toggle so the next changesets can implement them against an agreed design.

go-gost/x/socket/websocket_reporter_test.go

@@ -110,6 +110,14 @@ func TestSanitizeWebSocketURL(t *testing.T) {
 	}
 }
 
+func TestNewWebSocketReporterUsesReducedMetricInterval(t *testing.T) {
+	reporter := NewWebSocketReporter("panel.example.com:443", "abc")
+
+	if reporter.pingInterval != defaultMetricReportInterval {
+		t.Fatalf("expected metric interval %s, got %s", defaultMetricReportInterval, reporter.pingInterval)
+	}
+}
+
 func TestFormatWebSocketDialErrorIncludesHTTPStatus(t *testing.T) {
 	err := errors.New("websocket: bad handshake")
 	resp := &http.Response{