How can whole systems think: collective intelligence assemblies and the case for ‘intelligence design’