I'd love to see some statistics for success on that 'weld nut onto a broken bolt' method. You're trying to weld something on that has a smaller ID, so a small weld, usually in a bad location to really see what you're doing. Then you're heating and changing the metallurgy at the tip of a fastener that's seized enough along its length to break in the first place. I've tried it a few times and the welded item always twists right off. Might work if you heated the surroundings cherry-red, but probably requires acetylene, I doubt MAPP will get you there on a cylinder head -- it's too large a heatsink.
My method:
For non-critical fasteners (such as a manifold stud and VC bolt like you're dealing with) I never use any type of extractor. As mentioned, they are a harder metal, so that they bite in, but then the harder metal is brittle so more likely to snap, and then you're hosed.
I always grind them down flat if possible, use a spring-laoded center punch and tap the center of the bolt as best I can, then start small and drill successively larger with LH drillbits. Theoretically there's a chance that the LH drill will back the bolt out, although it's never actually happened for me yet (duh, because the bolt is seized!). I drill larger up to the tap size, then re-tap it.
I rarely am drilling perfectly on-center, so when I re-tap there's a mix of old bolt in the threads, so I always install a stud in afterward, and would not do this for a critical fastener (like a headbolt), or for a R&R bolt (like the long alternator bolt). But for VC bolts, exh studs, T-stat bolts, it has worked well for me as a near-permanent repair numerous times over the last 20 years.
As for worrying about hardened seats -
I drove an untouched 68 318 from 110k to about 180k at mostly highway speeds from 1997 til maybe 2005. Near the end it was starting to run a little rough (probably valves, although I never did check). So although there is a vulnerability there, I wouldn't just automatically install $3-400 worth of seats unless it was part of my long-term plan for the car. As someone mentioned, with the low-tension springs of a factory non-HP cam, and with lazy 70s highway gears, you're not beating the seats as much as a hotrodder would, so I don't think it's the best place to spend your money. It sounds like this car is new to you? If so, get to know it a bit first and figure out a long-term plan for it instead of just automatically 'doing it the right way' like the internet tells you. Other people will always spend more of your money than necessary. Another thing to consider is that re-working everything on the heads seals them up better, with slightly higher CR, and you'll be putting them on used rings so there's a chance to burn a little oil afterward. Maybe not a big deal on a low-CR 70s engine, but...
