"For tornado and earthquake proof, what magnitude on their respective scales is the design supposed to withstand?"
The problem with PWR designs is that they're a _giant_ pressure vessel, filled with _tonnes_ of water at 400C and 20-40 atmospheres and the nooclear stuff is in the middle.
That makes them akin to a balloon waiting for a pin.
Steam explosions are bad enough but when there's the added complication of loose radionucleides getting into the biosphere AND the fact that the natural temperature of fission reactions is around 11-1200C AND that water is corrosive when it's extremely hot and under pressure (not to menion the fact that nuclear plants add boric acid), you have a problem when the water gets out.
That's why the buildings are so big (containing the steam) and there are so many backup systems to make sure the reactor never runs out of water.
Other designs would use coolants which don't need to be pressurised around the hot stuff (which changes the entire protection design requirement), wouldn't boil below the natural temperature of the reaction (so don't need pressurisation) and wouldn't let water get in contact with radionucleides.
Thats the premise of using gas, lead, sodium or molten salt cooling, but let's be honest - using a metal that burns furiously in contact with air isn't the brightest idea (but the USA, UK, Japan and Russia have all done it and all found out why it's a bad idea - lookup "monju") and lead cooled fuel rods are a problem if they get cold as the russians found out the hard way on a couple of submarines.
Gas cooled systems work well, but the gas can still leak out (and has on occasion - UK AGRs).
Molten salt systems freeze at 300-400C so any leaks don't go far, the fuel is dissolved n the salt so you can chemically reprocess on the fly using flow reactions (not critical chemistry, you only need to clean it enough to keep the reactor running) and will happily sit there at 1400C without doing anything nasty.