The wire (deck pendant) sits proud of the deck in current systems.
As GK says, what is happening at / near mid-span of the cable is that the wheels running over the wire depress it close/onto the deck. Because the jet is doing something like 120kts relative (~60m/s) at that point, the time between the wheels depressing the wire and the hook point passing it is (2.5/60) or 0.04 secs. This is insufficient time for the cable to spring back up.
The trough-type arrangement you're describing would have to be very narrow (driven by the wheel track on both F18 and F35) and of the order of 1-2m in width at best, to avoid the same problem. However, by doing that you're automatically constraining the aircraft to recover within 0.5 to 1m laterally of the deck CL, which will increase the pilot burden / reduce operating envelope and also putting an osbtruction in the deck which in addition to being a corrosion / FOD trap would also be a hazard to tyres during a landing. The current safe parking lining allows for a 3m deviation off CL for the E2. It's a lot of trouble to go to before trying to sort the hook itself.
A new hookpoint is the obvious first thing and if that doesn't work, some sort of hook extension needs looking at, but that will be much trickier. If that doesn't work then they are in trouble. One thing is for certain, whoever in the project was responsible for it getting this far without the basic check on geometry for carrier landings, needs a shoeing of biblical proportions.