So I'm not going to work this all out as it's just algebra, but I believe the following is what you'll want to do.
$$L_z = \frac{\hbar}{i}\frac{\partial}{\partial \phi}$$
$$L_x = i\hbar \left(\sin \phi \frac{\partial}{\partial \theta} + \cot \theta \cos \phi \frac{\partial}{\partial \phi}\right)$$
Now since derivatives are involved, you can't just compute a simple commutator and get the result you want. You have to "act it" on a test wavefunction.
$$[L_z, L_x]\Psi = i\hbar L_y \Psi$$
If you can show that is true, then you have proved it.
My Solution
Here is the solution fully done out. The algebra really isn't all that complicated. I recommend getting a big whiteboard and writing everything out explicitly and just go one step at a time and don't get overwhelmed. There are no complex tricks here, just follow the simple rules of derivatives (there are a few chain rules involved however).
$$L_z = \frac{\hbar}{i}\frac{\partial}{\partial \phi}$$
$$L_x = i\hbar \left(\sin \phi \frac{\partial}{\partial \theta} + \cot \theta \cos \phi \frac{\partial}{\partial \phi}\right)$$
$$L_y = i\hbar\left(-\cos \phi \frac{\partial}{\partial \theta} + \cot \theta \sin \phi \frac{\partial}{\partial \phi}\right)$$
We now act on a test wavefunction (make a habit of this when computing two sides of commutators, and ALWAYS do this if derivatives are involved)
$$[L_z, L_x]\Psi = i\hbar L_y \Psi$$
Expanding the commutator on the left side, we get
$$[L_z, L_x]\Psi = L_zL_x\Psi - L_xL_z\Psi$$
$$(1) \to L_zL_x\Psi$$
$$(2) \to L_xL_z\Psi$$
For Equation (1) we have
$$(1) \to \left(\frac{\hbar}{i}\frac{\partial}{\partial \phi}\right) \left(i\hbar \left(\sin \phi \frac{\partial}{\partial \theta} + \cot \theta \cos \phi \frac{\partial}{\partial \phi}\right)\right)\Psi$$
$$\to \hbar^2 \left(\left(\frac{\partial}{\partial \phi}\right)\left(\sin \phi \frac{\partial}{\partial \theta} + \cot \theta \cos \phi \frac{\partial}{\partial \phi}\right)\right)\Psi$$
$$\to \hbar^2 \left(\cos \phi \frac{\partial}{\partial \theta} - \cot \theta \sin \phi \frac{\partial}{\partial \phi} \right) \Psi + \hbar^2\left(\sin \phi \frac{\partial}{\partial \theta} + \cot \theta \cos \phi \frac{\partial}{\partial \phi}\right) \frac{\partial \Psi}{\partial \phi}$$
$$(1) \to \hbar^2 \left(\cos \phi \frac{\partial \Psi}{\partial \theta} - \cot \theta \sin \phi \frac{\partial \Psi}{\partial \phi} \right) + \hbar^2\left(\sin \phi \frac{\partial^2 \Psi}{\partial \theta \partial \phi} + \cot \theta \cos \phi \frac{\partial^2 \Psi}{\partial \phi^2}\right)$$
For Equation (2) we have
$$(2) \to i\hbar\left(\sin \phi \frac{\partial}{\partial \theta} \cot \theta \cos \phi \frac{\partial}{\partial \phi}\right)\left(\frac{\hbar}{i}\frac{\partial}{\partial \phi}\right)\Psi$$
$$\to \hbar^2\left(\sin \phi \frac{\partial}{\partial \theta} + \cot\theta \cos \phi \frac{\partial}{\partial \phi}\right)\frac{\partial \Psi}{\partial \phi}$$
$$(2) \to \hbar^2\left(\sin\phi \frac{\partial^2 \phi}{\partial \theta \partial \phi} + \cot \theta \cos \phi \frac{\partial^2 \Psi}{\partial \phi^2}\right)$$
Taking (1)-(2), we get see that the second term from (1) is identically (2), so we get
$$(1)-(2) \to \hbar^2\left(\cos \phi \frac{\partial \Psi}{\partial \theta} - \cot \theta \sin \phi \frac{\partial \Psi}{\partial \phi}\right)$$
Which is now our left hand side. Computing the right side, we plug in $L_y$ to get
$$i\hbar L_y \Psi = (i\hbar)(i\hbar)\left(-\cos \phi \frac{\partial}{\partial \theta} + \cot \theta \sin \phi \frac{\partial}{\partial \phi}\right) \Psi$$
$$\to \hbar^2\left(\cos \phi \frac{\partial \Psi}{\partial \theta} - \cot \theta \sin \phi \frac{\partial \Psi}{\partial \phi}\right)$$
Thus we have equated sides and proved the theorem.