Add call_before_subproblem_solve callback in MindtPy

[ZedongPeng]

Summary/Motivation:

This Pull Request introduces the call_before_subproblem_solve callback in MindtPy, enabling customized warm-starts for solving NLP subproblems. This feature has been thoroughly tested by @jialuw96 and @adowling2.

Changes proposed in this PR:

• call_before_subproblem_solve callback in MindtPy

Legal Acknowledgement

By contributing to this software project, I have read the contribution guide and agree to the following terms and conditions for my contribution:

1. I agree my contributions are submitted under the BSD license.
2. I represent I am authorized to make the contributions and grant the license. If my employer has rights to intellectual property that includes these contributions, I represent that I have received permission to make contributions and grant the required license on behalf of that employer.

[adowling2]

I confirm I was able to run this code 2 weeks ago to reproduce paper results.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 88.45%. Comparing base (6393961) to head (e4920cb).
Report is 126 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3251      +/-   ##
==========================================
- Coverage   88.45%   88.45%   -0.01%     
==========================================
  Files         849      852       +3     
  Lines       95329    98139    +2810     
==========================================
+ Hits        84327    86808    +2481     
- Misses      11002    11331     +329     

Flag	Coverage Δ
linux	86.44% <100.00%> (+0.03%)	⬆️
osx	75.65% <10.00%> (-0.61%)	⬇️
other	86.63% <100.00%> (+0.02%)	⬆️
win	83.95% <100.00%> (+0.04%)	⬆️

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.

[emma58]

I'm confused that the docstrings say 'after' when I think the callback is 'before', but otherwise this looks fine.

[ZedongPeng]

Thank you @emma58 . I just corrected 'after' to 'before'.

[jsiirola]

This is fine, although it could be improved by adding a test that exercises the "new" option (adding the test will help ensure that we don't accidentally break the functionality in the future).

[ZedongPeng]

The callback test has been added.
The callback test is designed to make the algorithm cycle and terminate without finding the optimal solution, which is different from the case when the callback function is not invoked.