Another Update: The short version is that the repair is made, and I'm very pleased with it. I ended up doing pretty close to exactly what you suggested above. I wrote up an article in "Hunter 34 Modifications and Upgrades" titled "Port Water Tank Crack Repair" in case others want the details. The settee top now comes off the same way everyone else's does, and the area under there has been cleaned for the first time in 33-ish years. Ooh, it was ugly. Pretty now, though. I'm a happy guy!
I'm now halfway down the compression post rabbit-hole. I'm attempting something that I haven't seen others try, which is doing it with the rig still in place, but with all the standing rigging loosened right off. I figure the mast weighs nothing compared to the compression force, so as long as I support both sides it should be okay. The other thing I'm trying to do is to do it through the 5-1/2" x 3-1/2" hole that the old post cracked in the glass, without cutting away a big chunk of the underside of the beam. Most others have used pretty long beams under the mast step; as an engineer, I don't see the point of going much wider than the mast section itself, which is only about 5". That's where all the force is. So the plan is a beefy piece of aluminum, roughly 7" long, 3" wide, and 1-1/4" thick, held up by a new 2x2x0.25" aluminum square tube bolted to the underside. It will be a bit like building a ship in a bottle, but no FRP work or large trim pieces.
A couple of curious things: First, I'm cutting into the wood inside the beam, which is a really slow process through my little access hole, but what the heck. I've removed all the wood to the top of the channel directly above where the old compression post was, and in spite of the fact that the FRP bottom of the beam had buckled about 1/8", there is no sign of rot in the wood whatsoever. The wood actually smells like they just cut the tree down! This tells me that the compressive strength of the wood in the cross-grain direction simply wasn't enough. Consulting an old textbook, the compressive strength of douglas fir perpendicular to the grain is only about 330 psi (it's more than 10 times that in the grain direction). The bearing surface of the top of the compression post was about 3.5 x 5, or 17.5 in², theoretically good for just under 6,000 lb. On my boat, at least, the compression post was cunningly situated so that it wasn't under either of the vertical FRP sides of the beam, and the FRP underside of the beam was only about 3/16" thick. In other words, the cross-grain fir took the whole load. And clearly, since the theoretical design load is somewhere between 12 and 20,000, it didn't cut it.
The other odd thing is that although I haven't cleared 100% of the wood around the tube coming down from the mast, it doesn't look like there is any wiring at all in the transverse direction inside that beam. The wood seems to be solid across the whole cross-section. All the wiring from the mast goes straight down through a PVC tube through the fir and out the bottom of the FRP beam, then down inside the compression post and through the floor. There was some rot in the top of the compression post right around the wiring exit, so some water did drip down, but the post was fundamentally still solid.
Short version: you don't need rot to force you to make this repair. It wasn't strong enough even when the wood was intact.