Update. The block of wood between the piston and the head worked. With the head off, I ran a suitable branch through the table saw, cutting off a 1" section. Popped it between the head and piston and started gently swinging. After about 2 minutes of swinging, I noticed that the gib key slots were out of alignment, and gradually it started to move more easily - loads of oil helped. When it did come off, the mounting surface was not corroded, but it was huge - there was no way that was coming off by conventional methods.
One thing I have noticed is that the small end bearing is loose in the conrod. I have no idea whether it was loose before the flywheel swinging, or whether it was caused by the flywheel swinging, but it is worth checking. I would imagine that the impact of flywheel swinging is not a lot compared to thousands of diesel explosions.
The low compression was due to a blown head gasket, and the rings were a bit sticky. The bore is within tolerance and in good condition, so I will re-ring and deglaze. Judging by the inch of sludge in the crank case, the oil was never changed from new. Given that, the bearings are in remarkable condition, the big end is a bit scuffed, but the mains are clean.
One question: does anyone know how the main bearing at the flywheel end gets its oil? The bearing has an oil hole in it, but there is no corresponding hole in the cover, and even if there was a hole in the cover, I can't see an obvious feed route for the oil. The back (gear end) one is as expected, with bearing and feed holes aligned.
Edit: looking carefully at the cleaned up parts, I see an oil drilling in the flywheel bearing housing, and someone appears to have put the bearing in without considering this. There is an oil flow to the bearing, but it is partial. Pretty amazing that it is in such good order!