That's the "how". The other commenter was unsure about "when".

Fusion is still decades away from being commercially viable, so while continued funding for RnD for it should be important, it still has a long way to go before we can even start making plans for power plants operating on it.

As for Nuclear having long set up time... that's not a reason to not use it. That's a reason to hammer out plans and start finding ideal sites for it to service high density and high demand areas as soon as possible. The other half of this coin is that activists and legislatures need to take a serious look at Nuclear Regulations and cut the things that are simply not needed. Nuclear continues to be a viable option in spite of the severe restrictions placed upon it, restrictions that no other form of energy production could even be viable much less profitable. If we equalized these limitations and restrictions, you can pretty much guarantee that Nuclear would suddenly become #1 choice and still maintain the highest safety rating per watt produced.

Solar, Wind and Nuclear all complement each other very well, with the downsides of each generation being easily offset by the other two forms of power generation. Wind and Solar take a lot of space to make farms that produce enough power to feed a large city, where as a nuclear plant can produce the same amount of power with considerably less space. And the two can be interwoven together, with most nuclear plants already having 'no go' areas or restricted zones around them that could easily be filled in with Solar and Wind.

Solar will give you a set amount of power each day the shines, varying throughout the year but works as a good baseline. Initiatives could be pushed to add rooftop solar to most homes, businesses and other buildings or infrastructure that people don't need to walk across or stand upon, increasing the base line here in a relatively cheap and low cost way, though of course these panels do need to be cleaned regularly so just plastering them to the sides of skyscrapers isn't quite as viable as people like to think it would be.

Nuclear is a very efficient form of power generation with variable output. While it is true that Nuclear plants need some time to vary their output, this isn't nearly as big of deal as people make it out to be given the rate of power generation can be changed in a matter of minutes, not hours, and using historical data of power usage year over year and monitoring current conditions, a properly built and managed nuclear plant can easily adjust on a schedule to raise and lower it's output ahead of rising and falling demands throughout the day. And using modern reactor designs, these advantages become more pronounced with faster reaction times in generation and more thorough use of fuel leaving considerably less waste.

And Wind power acts as a great back up and buffer for all other forms of power generation, in rural areas wind power can easily provide 100% of the power needs and more to rural and suburban communities, and when combined with other forms of power wind can easily and rapidly bring turbines online or shut them off to follow the demand of the power grid and bridge the gap during any anomalous spikes and valleys in usage. Wind turbines do take a lot of space per gigawatthour produced, but in rural areas with farming or livestock where lots of land isn't really being used and developed for other reasons, wind turbines make sense, and with good a good power grid that can transmit power over hundreds of miles, can incorporate large areas of empty land into high density area in the same region to meet power needs with minimal pollution and carbon footprint.

Of course each power generation has it's problems, Solar only works when the sun shines obviously, requiring large scale power storage which isn't exactly feasible with the technology we have today. Wind turbines seem great on the surface, but their blades are made of carbon fibre in most instances, and have 7 year life spans at most, with no current programs or policies in place to break down these turbine blades that will not decay naturally, meaning roughly every seven years landfills and dumps get thousands more semi-trailer sized blades dumped into them with no long term plan on how to recycle or reuse them. And Nuclear power has the waste issue, and more importantly a PR issue. With older 60s and 70s designs, these reactors did generate notable amounts of waste, but this is less a problem with more modern reactors that can even run previously 'depleted' waste fuel through them to generate power while also reducing the amount of radioactive material left over. And of course the longer setup times, which addressed above is not actually something they should need, given the absurdly high standards placed upon Nuclear by skeptics and detractors decades ago mean many other power plants, factories and even government buildings would immediately be disqualified and turned into exclusion zones due the natural radioactivity of the building materials such as granite which we know to be harmless to humans.

It's going to take decades to decarbonize, the current Paris accord 2030 goal will not be met at all. Biden has set a target date for the US for 2050 to achieve net zero. That's 37 years away.

Currently we're going in the opposite direction:

"The Production Gap 2021 report states that world governments still plan to produce 110% more fossil fuels in 2030 (including 240% more coal, 57% more oil and 71% more gas) than the 1.5 degree limit.[118]"

People are talking about solving climate change as a sprint, but really it's a marathon. In fact, we don't even have the technology to fully decarbonize all sectors. It's going to take an "all hands on deck" approach and yes, will cost trillions of $$$ to achieve.

I'm pretty doubtful that we will achieve net zero energy and zero carbon before the end of the century. The first 50% of emissions reductions will be the easiest of the low hanging fruit; each successive % reduction will be that much harder and more expensive to achieve.

Tying back to your comment, each nuke plant permanently displaces millions of tons of CO2 emissions per year.

the fact we'd be trading dependency on resources from a group of unstable countries to a dependency on resources from another group of unstable countries

I'm not sure calling Canada (up until 2019 the largest exporter and miner of uranium) an unstable country. Unless you want to talk about our looming housing crash.