Fix timing issue in very_long_log_message test

  Move FlushAll() call to occur immediately after logging to ensure proper
  synchronization, and increase wait time from 5ms to 50ms to accommodate
  slower CI environments (looking at you GHA 🤪).

Author: wayneeseguin

pkg/omni/boundary_test.go

@@ -181,13 +181,14 @@ func TestBoundaryConditions(t *testing.T) {
 		longMessage := strings.Repeat("A", 1024*1024)
 		logger.Info(longMessage)
 
-		// Give message time to reach the dispatcher
-		time.Sleep(5 * time.Millisecond)
-
+		// Ensure message is fully processed by flushing
 		if err := logger.FlushAll(); err != nil {
 			t.Errorf("FlushAll failed: %v", err)
 		}
 
+		// Give additional time for any background processing on slower systems
+		time.Sleep(50 * time.Millisecond)
+
 		// Should handle large messages
 		content, err := os.ReadFile(logFile)
 		if err != nil {