Since my last update, I've actually driven the car (7 miles to my parent's garage but its a wee victory). It seems that the non-start problem was actually caused by over fuelling - the new plugs, leads, dizzy and coil appear to have strengthened the spark sufficiently to start the car (although it can still be touch and go - very easily flooded). The car now burns very very rich when its running. If it does flood, it can be started by disconnecting the fuel rail (hence depressurising the system), then reconnecting and cranking again. It seems that the small time it takes to pressure the whole system up is enough for the engine to catch.
So now need to work out what causes over-fuelling bearing in mind I've changed the coolant temperature sender and the manifold temperature sensor (the most obvious suspects) with no effect.
Although I can't remember the full list of MEMS inputs off the top of my head, the only other one that I haven't tried or replaced that would seem pertinent is the Lambda Sensor. I find it hard to believe that it could be so wrong as to increase the fuelling to this level. I would think it would go open loop long before. Or maybe open loop IS horrifically rich (?) but I kind of doubt this is normal...
Other considerations;
1) the fuel is (and was) always fine - I drained the tank and ran the Vectra off it just fine...
2) the new plugs are now getting so sooted up that I can see the starting problem coming back soon. Is there any way to clean plugs? (My dad reckons yes with petrol and rags but he's over sixty and in his day, everything could be repaired with petrol and rags. Or jubilee clips.
3) Compression is way low (but fairly even across the cylinders). I know there is oil getting into the bores when it sits for any length of time, but this isn't new and I can't see why it would suddenly effect the fuelling to this extent?
Getting very close to just sticking it into a garage and getting someone else to look at it. May need to mortage the house for that though...