Number Theory is basically the exploration of the deep consequences of a set of (ultimately arbitrary but apparently obvious) axioms. The exploration of those consequences arises from the fact that they are ultimately too plentiful, ramified, and incompletely provable (per Godel’s Incompleteness Theorems & Turing’s Halting Problem).
Isn't that true for all fields of pure math?
Number theory uses a much smaller set of axioms.
It should be stated that most mathematicians don't really mind the logical foundations of their work when they are actually working in the same way that most programmers don't worry about assembly language or transistors.
Number theory is actually an example of pure math having real world impacts despite its nature. It’s very much about studying math for maths sake - though there are also people interested in it for their own applications.
They’ve gone the brute force route with number theory because proving things in it is hard - in the sense that a hard number theory problem takes centuries to solve otherwise.
And yet the man who gave us physics ( newton ) also gave us calculus. Rather calculus was invented to help describe and model physics and the natural world.
IANAM, but I doubt there was ever a modern calculus theorem that was hinted by a brute force search. One can argue that Calculus is "made up" and thus research in this field is finding deeper logical consequences for the initial axioms, while natural numbers just "have properties" and research is Number Theory is trying to uncover these properties by brute force and later trying to prove them, arguably quite unsuccessfully so far. It's an interesting tidbit in the philosophical question of "Does math exist only in our head?"