I don't know about the specific grants, but in general many of these early career fellowships are intended for the applications to develop the necessary skills to become independent researchers. Grant writing is always a balancing act: on the one hand, you want to have interesting and challenging research questions (indeed, if you already have all the answers, why should they give you the money). On the other hand, as you rightly note, reviewers should be confident that the applicant has the skills (or potential) to complete the project succesfully.
In my view, this also applies to the "training" part of the grant. More specifically, I believe that they want to see which skills you hope to acquire, and how they will help you in your research project and future career. You should also be able to convince the reviewers that your plan is feasible. For example, for your example of learning how to program in C++, you could outline previous programming experience. Or perhaps your supervisor has succesfully used it in various projects, and will be able to help you learning C++. Or the university might offer training programs addressing this.
In short, it is expected that you may not yet have all the skills for the project, but you should be able to convince the reviewers that you have a well thought out (and realistic) plan to acquire them and bring your research project to a successful conclusion.