I'm trying to calculate the Haar integral on $SU(2)$ of a given function $f: SU(2) \to \mathbb{R}$.
For this particular function, I know the value of it on the subgroup
$$T = \left\{ \begin{pmatrix} z & 0 \\ 0 & z^{-1} \end{pmatrix}: z \in \mathbb{C}, |z|=1 \right\} \leq SU(2)$$
and that every element of $SU(2)$ is conjugate to something in $T$.
I happen to also know that $f$ is a class function, i.e. $f(A) = f(XAX^{-1}) \forall X$, indeed, I only have an explicit formula for $f$ on $T$, say $f\left( \bigl( \begin{smallmatrix} z & 0 \\ 0 & z^{-1} \end{smallmatrix}\bigr) \right) = \overline{f}(z)$.
So if I were to calculate the Haar integral $\int_{SU(2)} f(A)dA$, I'd ideally want to equate it to an integral over $T$ in some way, that is
$$\int_{SU(2)} f(A)dA = \int_{T} f(A') \omega(A') dA' = \int_{0}^{2 \pi} \overline{f}(z) \overline{\omega}(z)dz$$
Where $\omega$ is some kind of weight function quantifiying how common $A' \in T$ is a conjugate of $A \in SU(2)$ over all of $SU(2)$. That is, something similar to how the Jacobian acts during change of variables in multivariate integration. $\overline{\omega}$ is defined like how $\overline{f}$ is.
But I'm struggling to figure out how I might begin calculating $\omega$ (all I really need is to calculate $\overline{\omega}$) or if this is even the ideal approach to calculate the Haar integral.
You want the Weyl Integration Formula in the simple case of SU(2). This set of lecture notes seems to work it out nicely https://www.math.utah.edu/~milicic/Math_6260/weyl_character.pdf
Also the best elementary introduction to Haar measures on compact groups I know is the one by Joel Feldman https://personal.math.ubc.ca/~feldman/m606/haar.pdf the formula for SU(2) is in Example 5, ii.
Another approach which is suitable for integrating polynomial functions is explained in https://mathoverflow.net/questions/255492/how-to-constructively-combinatorially-prove-schur-weyl-duality/255853#255853 It is formulated for $SU(d)$ but it considerably simplifies if you plug $d=2$.