If I'm only working on one feature at a time then actually it's largely ok, it's when I have to manipulate the commit graph that things start to really go wrong for me. But rebasing I still don't understand, even after using git for over 10 years. I mean, I understand conceptually what it's trying to do, but every time I run `git rebase` I end up in a crazy wild state.
For example, suppose I'm working off main and I have a new branch right off HEAD with five commits on it. I realize that oh actually I want to port this onto a different branch that was HEAD~5 with three new commits that aren't on main. When I run this, git suddenly seems to merge conflict a whole bunch of stuff unrelated to my commits (and are related to the three commits not on main) and it gets worse the further from main the relevant branch is.
What I really want to do is essentially take the last five commit diffs and apply them to the HEAD of some other branch instead, like how you might drag and drop the graph nodes.
Currently to work around me not understanding why git doesn't align with my mental model, I un-commit my changes, git stash them, switch to the target branch, create a new branch, and then git stash pop. This seems to work much more reliably but it's so tedious to do every time that I kind of avoid doing this sort of change as a whole.