Overall a neat idea with plenty of scope to try out different features, although I recommend ASAP doing a process of game modelling as taught in the lectures to work out what is most feasible and to prioritise. For example a complex rewards/upgrade system may prove to be a lot of work, but you may be able to predict that work with a bit of research, and reduce the amount of work by sticking to just a small number of fun and interesting upgrades rather than too many. Balance between points and rewards strikes me as something that will be tricky to get right. Additionally an easy way to gain some time back might be to do fewer than 5 levels. Take a look at the Assignment 3C rubric to pragmatically choose which features to go with.
All-in-all looking forward to seeing how this one pans out, well done!