fix: make stepk match galsim

[beckermr]

This PR gets stepk for interpolated images to match galsim exactly. 🎉

xref: #189

[codspeed-hq]

Merging this PR will not alter performance

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

✅ 36 untouched benchmarks

---

_{Comparing stepk-fix (2f1bdb4) with main (88d4aec)}

[beckermr]

Test failures here are due to maxk not matching.

[beckermr]

Let's go ahead with this PR for now. I still want to look into maxk more, but this PR leaves the code in a pretty good working state.

[beckermr]

@ismael-mendoza this is ready for review!

[ismael-mendoza]

thanks! I can take a look today

[ismael-mendoza]

Just a few questions and some changes. If you think the first two are actually a bug, not sure if these should go into a different PR...?

[beckermr]

Let's see if removing this unused code works OK. I think it might and I do not recall why or what I was trying to do before.

[beckermr]

OK @ismael-mendoza! Thank you for the very helpful code review! This is ready for another look.

[beckermr]

@ismael-mendoza I recalled the general reason for the two classes here and added some notes on it.

[ismael-mendoza]

Thanks LGTM!