Assuming the cat run ends before the start of the ramp, as is needed for a straight cat. The aircraft nose wheel then has to detach from the cat before the start of shuttle deceleration. So the nosewheel is probably clear of the deck at this point, not rolling along it.. At some point it will then hit the ramp producing a retarding force at the nosewheel as well as an upward one. The aircraft needs to be designed for this.
That it has not been done is probably because aircraft tend to be designed to fit existing carriers and any carrier upgrades are compatible with existing a/c fleets. Plus, if you have most of your capability built around a conventional cat launch, why would you want to change the whole way you do business when it seems to be working?
Steam catapults work best when straight, so they are not compatible with a ski jump entry and continuation up ramp. I know little about electric ones, but they may be able to accept the bend needed to run up the ramp. In that case it is probably easier to design a curved cat than an aircraft that can be catapulted into the ramp entry and your aircraft then does not have to carry a penalty into action..
You also need to look at what you would get back . Yes, Ski jumps are dirt cheap; same price per pound as best bacon someone once said. Their benefit though is greatest for vectored thrust aircraft which can support some weight with thrust and the rest with lift whilst also accelerating on the ballistic trajectory from the ramp exit until lift is sufficient to support the whole weight. The alternatives to the F35B/ramp combination don't have vectored thrust, so don't make the same gains. I think you would need to look at the total a/c energy off a cat/ramp combination and compare it with the a/c energy off a straight, but longer, cat only launch to see whether there was any benefit.
N