The awesome power of exponential tipping points is not to be underestimated. When the cost/kWh curves intersect, it could happen fast. Look at the adoption of the Internet and smart phones for examples. Solar could go from <1% of our energy to >25% in 5-10 years. That might be conservative.
I'm seeing nuclear as a niche technology. In a post-fossil world I can see it for radionucleotide production (e.g. for nuclear medicine and space probes), maybe for a few really heavy industrial needs, and for some locations with very poor local renewable energy availability. But for the most part I see it as obsolescent.
I do think it might find its killer app somewhere else though: Mars. There are many reasons it would make more sense there.
Two examples: First, China doubled its life expectancy in a single decade, moving from one of the lowest to being at parity with the developed world. Second, in the early 1980s China implemented policies that moved them from having stagnant GDP per capita to steadily doubling it per decade, a pace it remains on since.
In fact, you could blame some of post revolutionary China's most disasterous events on the lack of bureaucratic inertia. The great leap forward was disasterous not just because the details of the plan were mistaken, but also because it was implemented so instantly and completely.
Anyhow, if you look at the data instead of repeating narratives that personify organizations, you'll see it's not quite so simple as saying that all Chinese economic decision-making can be summarily rejected as lumbering bureaucracy.
Renewables will only get so far if we don't stop increasing our energy demands exponentially every year. Since that seems unlikely, we will produce every single way we can.
You could cover 1/4 of the roof area in the US with solar panels and have enough power to close all other power plants. Add panels over parking lots and you get exponential demand back.