Rory Conlin

@unalmis can you look at adding the boozer stuff to `test_axis_limits.py`? It looks like we'll still need some special logic when integrating over theta/zeta for the `_mn` qtys.

> what changed in the master compute data, out of curiosity? The size of the boozer data. I think I mentioned in the meeting last week, master is currently wrong...

```python coil_sym = r"coils.rotatingellipse" field = MixedCoilSet.from_makegrid_coilfile(coil_sym,method='linear') plot_coils(coil_sym) ``` it looks like you're trying to plot the string `coil_sym` not the actual loaded coilset `field` (though we could probably make...

Might be useful when we want to do multithreading https://github.com/jax-ml/jax/issues/24756

> `test_jaxify` in `tests/test_utils.py` is failing on the CI for Python 3.9 with the error message: `ValueError: IO callbacks do not support JVP.` But it passes for me locally with...

> > change the other filenames to `integrals/_{*}_utils.py` etc. > > I made this change, but now in user code when making grids, I have to use: > > >...

isn't `fourier_pts` just `np.linspace(0, 2*np.pi, N, endpoint=False)`? and you can make a LinearGrid with an arbitrary number of points `LinearGrid(zeta=32, theta=64)` (also I don't think the power of 2 thing...

Also it looks like #1290 makes some changes/updates to the `Bounce2D` apis, would be good to do all of that here

After the fixes in #1043 this doesn't compile anything after the first iteration.

> ```diff > | benchmark_name | dt(%) | dt(s) | t_new(s) | t_old(s) | > | -------------------------------------- | ---------------------- | ---------------------- | ---------------------- | ---------------------- | > test_build_transform_fft_lowres | -1.56...