I'm a big fan of Python and just its standard libraries plus a couple 3rd party libraries depending on what I'm doing. Python is a great general purpose language. PyGame should be fine for basic video game stuffs, but I'd start simpler with just standard libraries.
VS Code is probably your best choice for an IDE.
It depends! Most AC to DC converters actually convert from AC to DC back to AC (at a higher frequency) then to DC again. The higher frequency means you can use a much smaller and lighter transformer for the same power. Depending on how it does the first AC to DC step, you could be ok giving it DC, but I wouldn't do it unless the power supply says it's rated for it.
XFCE usually. Two years ago it didn't support my weird multi monitor configuration, so I still have one computer with Gnome.
Here are a few ideas to help with motivation and project selection.
A lot of the time for me, my issue is that what I want to make is too open ended. I get stuck with decision paralysis. I keep trying to add features in my plan, or I don't know how I want to do parts of it because I don't know how I'll want to expand it in the future.
Artificially limit the scope of your problem. You want to make a game, but that could be a several year long indevor. You want a way to quickly feel successful in making something (for now). Possible limitations: Make a 2D black and white game out of only ASCII glyphs on a terminal. Make a game that can be won or lost in only 5 minutes of play. Make a game that only uses one or two buttons. Choose some or all of those.
Sometimes it can be nice to instead of starting from scratch, to extend someone else's work. Find an open source game in a language you know and add one feature to it. Add more only after the first feature works.
Reinventing the wheel can be fun. Pick a simple game and recreate it. Snake, Flappy Bird, and Pong come to mind.