There is indeed a cost v benefit element to all this. Yes a supercomputer could be used to mine bitcoin - but the cost of the supercomputer likely exceeds the benefits of mining the coins. So supercomputers will be put to other more important uses.
It's technically not even the cost aspect that makes a 'super computer' not a good Bitcoin miner, it's more the task area it's supposed to be super good at. It might be a super computer in one area but a complete sucker in some other areas. This is also why students thinking to utilize their super computers within the university network to mine Bitcoin won't come far at all.
Mining is such a waste of time. It's much easier and rewarding for people to just buy Bitcoin and sit on it. No stress about difficulty increases which eat through your income, no stress about broken gear, no stress about sudden changes in power rates, etc. Too many factors that point out how people should mess around with mining.
If you want to support the network, run a full node or provide liquidity to Lightning with a small chunk of your freshly bought coins.