Use chandralupta to find MPP in lambertw method

[cwhanse]

• Closes Use scipy's new optimize.elementwise functions #2497
• I am familiar with the contributing guidelines
• [ ] Tests added
• [ ] Updates entries in docs/sphinx/source/reference for API changes.
• Adds description and name entries in the appropriate "what's new" file in docs/sphinx/source/whatsnew for all changes. Includes link to the GitHub Issue with :issue:`num` or this Pull Request with :pull:`num`. Includes contributor name and/or GitHub username (link with :ghuser:`user`).
• New code is fully documented. Includes numpydoc compliant docstrings, examples, and comments where necessary.
• Pull request is nearly complete and ready for detailed review.
• Maintainer: Appropriate GitHub Labels (including remote-data) and Milestone are assigned to the Pull Request and linked Issue.

Uses 'chandralupta' when scipy>=1.15. Existing tests are good enough.

[cwhanse]

Restart of #2567

[cwhanse]

Pursuing the test failures, one condition tested is photocurrent=0. For that input, i_mp=nan is returned when the scipy method is used (#2567 (comment)).

nan is returned because the convergence fails on the first iteration, due to violation of the bracket condition. Quote from the scipy documentation for find_minimum:

requires argument init to provide a three-point minimization bracket: x1 < x2 < x3 such that func(x1) >= func(x2) <= func(x3), where one of the inequalities is strict.

Here, init = (0., 0.8*v_oc, v_oc) and func=_vmp_opt in the PR.

Evaluating at photocurrent=0. and v_oc=init, I get _vmp_opt(init, 0., ...) = array([-0.00000000e+00, 2.18952885e-48, 2.73691106e-48]). Note that the negative power is actually positive. The func values at this bracket fail the check.

Digging deeper, _lambertw_i_from_v(init, 0., ...) = array([0., -4.1359e-25, -4.1359e-25]) the negative currents are the root of the problem.

We could set small negative current to zero in lambertw_i_from_v But the MPP with chandralupta would still fail since photocurrent=0. since func(init) would be array([-0., -0., -0.]) and one of the inequalities has to be strict.

I think intercepting small photocurrent and skipping the MPP (or entire IV curve) calculation is better.

Your thoughts?