It's a problem all hobbyist face sooner or later. It might be a days, weeks, months, or even a few years on a project, but sooner or later the new idea fades and other shiny untested ideas shine.
As far as I know, everyone struggles with it to some degree, and it seems there is no remedy to "skip" it. There are strategies to help, and things that can make success more likely, but you can't skip overcoming this struggle:
Designing games is quicker and more fun than making games.
Don't get me wrong, there is tons of fun to be had in making games, but you will have to push through hard problems. You will have to times where you tread water for weeks at a time trying to figure out how to navigate to a solution. And the shiny new ideas will always be there tempting.
My best advice is to start incredibly small. A simple game that you can finish in a week. Maybe it has one level, maybe it has a few. Maybe it's part of a game jam. But actually finish it. Resist the urge to try something else, resist the urge to quit. It's going to be a bit of trash that in later years you'll want to bury, but that's okay!
Now make a totally different game that's two weeks long, but as you do, recognize when you are reusing strategies or when the work on your first project could be reused. And when you do, start storing that code up in reusable modules you can just attach to give behaviors to things. In this way, not only will you be forced to learn new things, because it's a new game concept with new challenges, but you've also started to build a library in your mind of how to solve various problems and a library of behaviors you can use to skip forward in the development process. This helps speed up the pace of going from idea to prototype.
Do this a few times. At some point, you graduate from building micro games to building prototypes. When you have a flash of inspiration on a game, spend a week or two making a prototype. It won't be finished, but it will let you try out that idea and find the ones that are fun.
And then you graduate to vertical slice. Spend a few weeks to three months making a polished vertical slice, something you would actually be proud of, even if it is still no where close to a complete game.
Now go back and look at your first few little games, and see how far you've come. If that doesn't give you the inspiration to see what you can achieve, I don't know what will.
This Godot 3.X video might have some of the components necessary: