Fission is a bad complement for solar power. A good complement for solar provides dispatchable fill-in-the-gaps type of power. However, nuclear power plants cost about the same amount of money per hour whether they're running or not, so are poorly suited as a solar complement.
You can get to 100% carbon free by using a combination of overprovisioning, source diversity, geographic diversity, storage and statistics. You can't get to 100% but you can get to an arbitrary number of nines and say "good enough". The grid is only 99.99% reliable so having generation be better than that has little value.
If you really want 100% reliability out of a primarily-solar grid you choose something the opposite of nuclear: something cheap to build but expensive to fuel. For example, synthetic natural gas. The cost of the fuel has little relevance when it only has a duty cycle of 0.1%. And they're cheap to build and likely don't even need building because we already have lots of them.
China built a lot of nuclear reactors in the 2010's, but is slowing their build rate. 10GW/year of nuclear power is not very impressive in a country that built 100GW of solar in the first half of 2024.
That being said, I still support restarting TMI. The main costs of nuclear are in building and decomissioning them. We've already incurred the cost of building TMI and we're already on the hook for decomissioning it, so running it is likely cost-effective once you remove the sunk costs.