People want their vehicles predictably usable, which is why any idea dependent on using car batteries as actual power smoothing will fail.
Also, as batteries can be charged wherever and whenever there is electricity, we can simply install PV enough for the daylight hours to produce the energy desired for both daylight and twilight/nighttime and then just charge daytime and discharge nighttime even though it's probably the other way around wherever you are because currently demand is generally lower relative to supply at night and therefore the electricity is generally cheaper at night.
And the same batteries can easily be given a second life, after degrading too much for car use and before refurbishment/recycling to make them suitable for car use again, as fixed batteries.
And again, this is just presuming only PV, not wind which reduces the variance in supply, and not any of the other things I listed which don't get affected by weather at all.
This is the thing people tend to forget. Even if a battery is "useless" for an EV, meaning that it has degraded to 60-70%, it still has 25-50kWh of usable capacity as-is.
And that's enough to run a house for 12-48 hours depending on how power-hungry your appliances are.
This isn't why.
Batteries stabilise grids. If every car was an EV, the batteries in the vehicles would be sufficient to stabilise a completely wind/solar grid. Heck, put PV on car bodies and you get 80-90% of their average demand, and that's one of the worst places to put PV.
Also, geothermal is not at all volatile, tidal is regular enough for a smaller storage requirement, hydro doubles as storage, and even with PV and wind larger grids are always less volatile and yes you can go global on paper.
No, the reason why grids aren't already renewable is that it takes time to build new infrastructure, combined with renewables only getting cheap enough to seriously consider that anywhere in the last two decades, and in most places in the last one decade.