Awfully convenient premise for your argument that the 297 kids in question who died from COVID we're—of course—high risk and would died anyway, and conversely, the 300 kids with heart problems were otherwise—somehow—the pictures of health. Even though the data provided doesn't suggest that.
...and that that increased risk factor of the kid COVID deaths—even if true—mitigates a 5.5x difference in sample size.
It appears that your difference between "so many" and "a bunch of" and "the vast majority" and the actual data is an ideological lens, not a statistical one.
OR
"My healthy kid is at small risk of COVID, but I’ll get them vaccinated despite the even smaller (by orders of magnitude) risk of serious side effects” is a reasonable position.
Of course, that reality-based conclusion wouldn't satisfy the ideologically-driven need to demonize the contrary position, so the baseless projection was added on as a bit of outgroup bias seasoning. It's an emotional argument, not a fact-based one.