Respect prob.bounds in all solver wrappers

[missing-user]

Only the constrained_solve overloads respected the problem bounds, the least_squares_solve overloads ignored them, unless passed with bounds=prob.bounds explicitly. This is unintuitive, since these functions are specifically overloads for LeastSquaresProblem, so I would expect it to use all the properties set on the object.

I suggest harmonizing the usage of constrained_solve and least_squares_solve to in both cases respect prob.bounds (as implemented in this PR), or at the very least raising a warning message to users, if prob.bounds were set, despite being ignored (I will adjust the PR if required).

[missing-user]

I also replaced print with logger.info in two places that stood out to me.

[codecov]

Codecov Report

Attention: Patch coverage is 60.86957% with 9 lines in your changes missing coverage. Please review.

Project coverage is 92.64%. Comparing base (d6bea2b) to head (adb4dec).

Files with missing lines	Patch %	Lines
src/simsopt/solve/serial.py	58.33%	5 Missing ⚠️
src/simsopt/_core/optimizable.py	0.00%	4 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #463      +/-   ##
==========================================
- Coverage   92.66%   92.64%   -0.03%     
==========================================
  Files          78       78              
  Lines       14514    14529      +15     
==========================================
+ Hits        13450    13460      +10     
- Misses       1064     1069       +5     

Flag	Coverage Δ
unittests	92.64% <60.86%> (-0.03%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[missing-user]

Replaced deprecated np.product with np.prod for numpy 2.0 support

[landreman]

Looks good to me. I think the failing tests are unrelated to the changes in this PR. Let's see if we can fix those in master. Anyone have ideas how to resolve those?

[missing-user]

Looks good to me. I think the failing tests are unrelated to the changes in this PR. Let's see if we can fix those in master. Anyone have ideas how to resolve those?

Thanks, it also seemed to me that way, tried looking into the issues with pyoculus myself but got sidetracked unfortunately.