On The Potential of Jointly-Optimised Solutions to Spoofing Attack Detection and Automatic Speaker Verification