By far the best way to do it is outside your interface. I use a spare headphone output on my digital mixer and there will always be one on most interfaces as well.
The best way to do it is to sum the L and R channels in a small passive mixer perhaps and feed a single channel amplifier and a single mono speaker. The speaker can be smaller too eg Auratone type size. NOT your two main speakers in mono. That is nowhere near as effective as one speaker. Mono means
one not two producing the same output. Even two speakers producing mono will give you some stereo imaging due to the nature of sound arriving at both ears at slightly different times. It adds to the mono confusion.
A different mono speaker also increases the number of systems you are hearing your mix on which is never a bad thing. And a small speaker shows up all sorts of mix issues too. It is also good down at low volume too but that is another story I guess.
The reason why this works is you never have to put anything on your masterbuss or switch it into mono etc. It is just a simple matter of turning up the headphone level feeding the mono system.
All your sounds now will line up behind each other in the small mono speaker. If they are too similar in sound and musical parts then they blend and will not sound separated. If you want to similar parts to be more separated then making them that way in a mono speaker goes a long way once they are panned etc..in stereo.
Any wide chorus effects etc may also sum badly in mono and therefore you will know you have to rectify that by possibly doing polarity reversals on one side and checking again etc.. Some very wide synth sounds don't collapse well in mono either alerting you to make adjustments etc..Using tight delays to create stereo width (which IMO is a slack thing to do) often performs badly in mono too. eg comb filtering etc..
You never know when you music will be summed to mono. It happens all the time. Be ready for it. Never just assume your mixes will always be heard in a wide stereo situation.
You can also use it very creatively too. I have heard some great mixes that start out in mono (and sound great) then they just smoothly spread out to stereo over time. They can jump back into mono for a short period etc..