You understand it wrong. Do-overs is not what I want. I want to do a lots of fancy do-it-mad-way things at certain points and then summarize all experience and after all do what reasonable and live with consequences. In real life you learn all your life, all those years, and there are a lot of people who help you, and luck/quantum effects. But in game you have only limited time to learn what possible. So it's not fair to make player do reasonable things without getting a lot of back knowledge.
First of all, i hope this answer doesn't read hostile. It is really not my intent here to be contentious. I want more to offer solutions or explain why the game developers/ audience like or dislikes something.
Also Metagaming is playing outside the boundaries of a game. For example in Dungeons and Dragons, metagaming is using the knowledge that your character CANNOT have (Which is probably what you refer to here). Playing over and over again is not metagaming in a roguelike because it is intended as the playstyle used.
If you think in RL you get the tools to do something, then Google solutions... From what i hear here, i would say that what you want here is more a tutorial rather then saves with that explanation. Also if you do something experimental IRL, you have also a chance to fail and do your work over again. Sometimes with lifelong consequences. (Open up a company that fails as a example. You can gather the knowledge how you open one. But you have consequences if you fail)
My personal advice for you is: Play as you wish. Mod the game if it doesn't pleases you. And if you do not want to mod/ cannot find mods or not develop own, play another game if the issues are a deal break for you. There are plenty of games out there which targets different audiences ^^ (Also i personally think save scumming by replacing files is one of the alternate solutions to replace a save function)
Roguelikes are intended for people who like games with consequences of loosing or making good/bad choices. Experimenting is a wager of risk/ possible reward there and if you are ready to deal with the outcome.
And for the devs i also have a personal point: Make the game you want to make. That creates the creativity in game development. Making a game with an own characteristic can be a risk, but i find these more interesting then the non-risk-cooky-cutter games that we often see today from less risking companies. (CoD is a good example. These games are mostly the same because they do not want to take the risk that people dislike it. So they stick with what they think people know and are familiar with)