You're mixing up two values here. Switzerland imports less than 10% of its electricity, the number you're probably referring to is the total of "Energieträgern", so oil, oal, gasoline, etc.

https://www.bfs.admin.ch/bfs/de/home/statistiken/energie/versorgung.html

googles

It looks like that's true, but that the ones returning the warmer water rather than evaporating it are also ones on the rivers (rather than the coast).

I don’t think you’d want to tie your ability to cool the reactor to outside air temperatures. Bodies of water do warm up, but are generally a much better “ultimate heat sink”.

Some places use the waste heat for district heating, in which case the waste heat isn't really a waste any more. I don't know if France does that, though. The Soviet Union was super-big on district heating. That won't consume all of the heat, and it really just gets you cheap heating for the city -- like, a large power plant will generate more heat than you actually need to heat a city.

In the US, we historically had some very limited district heating, though that was driven off waste heat from fossil fuel power plants, where you can do the same thing.

If you can make use of the heat output directly, it's an area where thermal options (fossil fuels, nuclear, geothermal, solar thermal) compare particularly favorably to non-thermal options (solar panels, wind, hydroelectric), because there the thermal options don't need to see the efficiency reduction from conversion to electricity and back.

googles

Sounds like they aren't currently using the heat for district heating (or at least some aren't), so they could probably use at least some of the heat some of the time for that.

https://www.sciencedirect.com/science/article/abs/pii/S0301421518300533

While being source of low carbon electricity, French nuclear plants also produce excess heat currently rejected to the surrounding environment. This heat represents approximately two thirds of the total energy generated (Safa, 2012). The whole French nuclear fleet and most nuclear reactors operating today in the world (277 out of 438) and tomorrow (59 out of 70 under construction; IAEA, 2015) are Pressurized Water Reactors (PWR). Similarly to cogeneration plants based on fossil fuels or biomass, PWR can be designed to produce both electricity and heat without affecting the reactor’s safety (STUK, 2009: p. 6). Doing so would involve reducing the amount of heat which is converted to electricity (Carnot efficiency). For the temperatures useful to conventional 3rd generation DH networks (80–120 °C), the thermal energy recovered is six times higher than the electricity losses (Safa, 2012). Production of heat for DH application has been experienced with at least 51 nuclear combined heat and power plant (NCHP; Leurent et al., 2017), hence providing valuable feedbacks for optimizing such energy systems.

That being said, the time when anyone cares about water flows and river temperatures is presumably the summer, and that's also going to be when there's less demand for heat. So it's more a way to get heating more-efficiently than an alternative to dumping heat somewhere.

Unless you're gonna create some absolutely colossal thermal energy store that could span seasons, like, put a cover on a lake, heat it up in the summer, and then use that heat in the winter.

You’re concerned about a cooling tower that literally just evaporates water? You realize those aren’t nuclear specific?

These guys say that there hasn't been a much research on microclimate effects since the 1970s, though they did look at it, and found that it slightly increases crop yields in the area in the US by increasing precipitation a bit.

https://www.cambridge.org/core/journals/journal-of-agricultural-and-applied-economics/article/effect-of-nuclear-power-plants-on-local-crop-yields/5CE7792374CCEF73CCBA9FC39BF131F6

The growing prevalence of clean energy raises the question of possible associated externalities. This article studies the effects of nuclear power plant development (and, as a result, the increased amount of water in the atmosphere from evaporative cooling systems) on nearby crop yields and finds that an average nuclear power plant increases local soybean yields by 2 and corn yields by 1 percent. Considering the low elasticity of demand for these crops, the yield increases translate to annual net benefits of $229 million (2020 US dollars) – $317 million in losses to farmers and $546 million in benefits to consumers.