The amount of risk present when casting a spell is entirely separate from how frequently a character is able to do something magical - so that's not an "option 3" so much as it is "option 1b" and "option 2b".While this post is phrased pejoratively, I think it's generally right. The better way to say it, I think, is that it's bad design to make a game whose core premise is that half of the players feel useless at any given time. As D&D has matured and the designers have gotten better at their craft, the later editions are designed so that everyone can feel powerful throughout the life of the game.
Now, I think that's all well and good, but the problem with the Potterization of D&D is the option missed by this post:
The third option being:
3) Let characters that focus on magic use (i.e. wizards) use magical means as often as they want, but magic is fickle and dangerous to use.
The problem I see with a lot of the magic in D&D is it's 100% good at what it does. There's no risk in dropping a fireball. A +3 weapon is just better than a non-magical version. There's no risk or opportunity cost. There's no chance of loss or damage. As a result, magic becomes humdrum because there's no reason not to use it.
Of course, my experience of D&D players is that, when it comes to D&D, most don't actually want to have risk added to their spell casting (check the lack of popularity for the Wild Mage for evidence).
And while you say there is no opportunity cost, I don't find that to be correct - to gain anything magical, something else is not being gained. Whether that is gaining spellcasting as a class feature rather than any other class feature because you chose to be a wizard and not some other class, or it is a DM giving you a +3 weapon rather than some other form of treasure, or even just using that specific +3 weapon (let's call it a longsword) rather than some other weapon (like a polearm, for example).