In terms of single-incident, I believe hydro takes the lead due to the Banqiao Dam failure (floods over 12,000 square kilometers killing ~30k directly then ~100k more from water contamination and famine). But even that's still dwarfed by the more gradual impact of climate change on the planet from fossil fuels.
Dam failures absolutely destroy vast areas and can contaminate them with oil and gasoline by flooding settlements. In fact hydroelectric power has killed a surprising amount of people.
Suppose you have a devastating earthquake that wrecks renewable power sources in an area. Bad part is you're out of power for a good while. But the second-order risks are very low - maybe some people are killed by falling wind turbines or solar panels, but there isn't any equivalent of radioactive contamination making the place uninhabitable for years.
Banqiao Dam failure alone flooded over 12,000 square kilometers, killing ~30k directly then ~100k more from water contamination and famine.
Even solar still requires various minerals the extraction and processing of which has environmental impact - like tailing ponds of toxic substances that frequently leak, particularly from earthquakes and extreme weather.
But to be clear, the human and environmental impact of solar, wind, nuclear and hydro pales in comparison to that of fossil fuels.
I specifically made my comment about wind and solar. I am not a big fan of large scale hydro for just this reason, as well as the impact on aquatic ecosystems.
> I specifically made my comment about wind and solar
You said "Suppose you have a devastating earthquake that wrecks renewable power sources in an area" - hydropower is considered renewable. I also did address a similar risk of solar (barely anything compared to fossil fuels, but same applies to nuclear).
Focusing on deaths is a nice trick of the nuclear industry because the largest negative impacts of their disasters and even non-disaster costs of operation are still terrible but do not manifest as immediate deaths.
This isn't just counting immediate deaths - else it'd just be ~30 for Chernobyl and ~0 for Fukushima.
In fact, Fukushima's death count seems to primarily be down to that "people had died indirectly as a result of the physical and mental stress of evacuation" - which I feel should at least partially be put on the tsunami that killed 20000.
Nothing is 100.0% safe - but as far as I can find nuclear comes incredibly close[0] even when including Chernobyl and Fukushima.
[0]: https://ourworldindata.org/safest-sources-of-energy