Fix estimate_parameters for binned and non uniform axis

[LMSC-NTappy]

Description of the change

Rebase of PR #2740 with the correct version of non_uniform_axes.
Fixes #2739.

Summary: this PR fixes incorrect behaviour in BaseDataAxis.value2index() which was not accepting arrays as input and was computing the axis gradient even for non-binned axes.

Progress of the PR

• Change implemented (can be split into several points),
• update docstring (if appropriate),
• add tests,
• ready for review.

[codecov]

Codecov Report

Merging #2743 (9574ada) into non_uniform_axes (a737190) will decrease coverage by 0.13%.
The diff coverage is 81.30%.

@@                 Coverage Diff                  @@
##           non_uniform_axes    #2743      +/-   ##
====================================================
- Coverage             78.02%   77.88%   -0.14%     
====================================================
  Files                   203      203              
  Lines                 31607    31144     -463     
  Branches               7020     6807     -213     
====================================================
- Hits                  24662    24258     -404     
+ Misses                 5130     5086      -44     
+ Partials               1815     1800      -15     

Impacted Files	Coverage Δ
hyperspy/_components/volume_plasmon_drude.py	47.05% <ø> (ø)
hyperspy/misc/array_tools.py	76.35% <30.00%> (-7.25%)	⬇️
hyperspy/axes.py	90.10% <62.50%> (+0.10%)	⬆️
hyperspy/_components/arctan.py	66.66% <100.00%> (ø)
hyperspy/_components/bleasdale.py	100.00% <100.00%> (ø)
hyperspy/_components/doniach.py	94.87% <100.00%> (+0.13%)	⬆️
hyperspy/_components/eels_arctan.py	89.47% <100.00%> (ø)
hyperspy/_components/eels_double_power_law.py	96.42% <100.00%> (ø)
hyperspy/_components/error_function.py	81.81% <100.00%> (ø)
hyperspy/_components/exponential.py	82.60% <100.00%> (ø)
... and 27 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update a737190...9574ada. Read the comment docs.

[ericpre]

This looks good to me and I have left a couple of comments.
I will leave someone else to review/merge it since I influenced this PR!

[jlaehne]

Looks good on first sight, I'll have a closer look in the next days when I find some time.

[ericpre]

This looks good to me, just need a tiny bit of tidying up. @jlaehne, I leave it to you. Thanks!

[jlaehne]

Thanks a lot for the help with the nua branch! I left a few questions, but as said previously this looks good otherwise.

[jlaehne]

OK, I merge this PR as the main objectives are more than covered. Further optimization that we discussed at several points can be contributed as new PRs. Thanks again!

[LMSC-NTappy]

This was a nice ride. Thank you!