It's different. Both from the technical side of things and the business case/use case side of things.

The business case side of things:

Consumers who want to take a vacation but are open to destinations. Other than co-terminal situations, generally speaking consumers depart from their local home airport.
Showing them the destinations along with the price point encourages the sale. The destination view can also include paid promotion that display certain destinations/airlines more prominently.

The opposite function that you're looking for has a much narrower use case.
I can see why a Flyertalk would want it - (Find long haul international deals in Premium cabins, and willing to pay for a cheap domestic /near-by EU reposition flight).
But that's a much smaller number of users and almost no promotional revenue opportunity. Airlines certainly won't pay to promote themselves if not downright hostile with this format.

For technical side of things -
A One-to-Many relationship is easier to cache than a Many-to-Many relationship. (Airfares varies by point of sale for the same trip)
Obviously, it is doable. Just that it is more costly computationally.

Expert Flyer or another tool marketed towards frequent flyers might have a business case to do it, just like the various 3rd party award ticket searching tools. But not a priority for Google Flights or a mass-market OTA.