I am having trouble calculating the following integral:
$$ {\int \sqrt{c-\cos(t)} dt} $$
where $c$ is a constant.
In the very specific case where $c=1$, it is rather easy, but I cannot seem to be able to generalize that.
I tried running it with Wolfram but it produces higher-than-my-paygrade math (example) about elliptic integrals and I can't help but think it should be easier than that.
No, it is not easier than that. Wolfram showing you "weird" functions is a sure sign, if not a proof*.
The elliptic integrals have become very classical. https://en.wikipedia.org/wiki/Elliptic_integral
It is easy to understand why the case of $c=1$ is simpler. In the plots you see that the blue curve is a (rectified) sinusoid, while the others are not, as can be checked analytically.
*A proof is possible but highly technical, based on Liouville's theorem. https://en.wikipedia.org/wiki/Liouville%27s_theorem_(differential_algebra)
Also note that a simple integrand rarely means a simple antiderivative. Try
integrate 1/(x^5+1).