A bit more on looking at the solution and failure diagrams
You need to know why a mistake fails, right?
The idea of reinforcement is basic to learning, as is the idea of imitation. Take the latter first. Looking at the solution gives you something to imitate. Imitation can take you pretty far, but obviously is not enough for a game as difficult as go. Still, it is important, and not to make use of it by not looking at the solution would be a shame. As for reinforcement, there are two kinds, positive reinforcement and negative reinforcement, related to reward and punishment, respectively. A game of go provides both: positive reinforcement if you win, negative reinforcement if you lose. One problem, OC, is the question of which among the typically more than 100 moves that you made in the game are good, and should be positively reinforced, and which are bad and should be negatively reinforced? Not always obvious.
For problems, solution and failure diagrams provide positive reinforcement to good moves and negative reinforcement to bad moves. Those are very important in the learning process. You ought to look, if only for the satisfaction of verifying your reading (positive reinforcement).
Now, for reinforcement to be effective, you need something to reinforce. Just looking at the solution may invoke imitation, but is useless for reinforcement. You have to work on the problem first. If you are pretty sure that you have solved the problem, then might be a good time to look at the solution and failure diagrams. But what if you are stuck? How long do you spend reading out a difficult position during a game? 1 minute? 2 minutes? 5 minutes? 15 minutes? 1 hour? I expect that there is a sweet spot, but I do not know what it is. As a dan player I was willing to spend 15 minutes, but I think that much time would be counter-productive for most SDKs.
Remember dithering? Going back and forth between apparently unacceptable alternatives? One problem with spending too much time on a problem is that you end up essentially dithering, by repeating unsuccessful lines of play while looking for alternative plays. If you are stuck, you are not finding those good alternatives. But by that process you are positively reinforcing bad plays. Reinforcement is not just about reward and punishment, it is about strengthening neural pathways. Every time you repeat a bad variation, you are positively reinforcing it, even if it leaves you feeling unsatisfied at the end (negative reinforcement). Positive reinforcement is more powerful than negative reinforcement. Taking too long on problems can build bad habits of thinking about go.
Chess grandmaster Kotov advised in calculating variations to calculate each branch of the game tree only once. Obviously, that avoids dithering. That advice has generally been rejected for actual play. However, as a discipline in solving problems I think that it is excellent. You choose your candidate moves at each turn and read each branch to whatever depth seems right at the time (or to the limit of your ability), and if you find no solution, you are done, and you have not reinforced any bad moves.
Now you are ready to look at the solution and failure diagrams.
Edit: Having considered this some more, I see that I ignored the role of the subconscious in solving problems. If you have spent time and energy trying to solve a problem, it can help to leave it for a while for your subconscious to work on. Overnight is not a bad period of time. I have sometimes woken up with the answer to a problem or question that I had been working on.