FIX-ENH update cutoff_at_kV behavior for bruker.py reader

[sem-geologist]

Description of the change

Changes how bruker.py reader reacts to cutoff_at_kV:

• None as value does no more any cutoff
• introduced possible value "zealous" which cuts at last nonzero value
• previous behavior with None can be activated with value "auto", with updated fallback if hv=0

This partly fixes the #2898

Progress of the PR

• Change implemented (can be split into several points),
• update docstring
• update user guide,
• add an changelog entry in the upcoming_changes folder (see upcoming_changes/README.rst),
• Check formatting changelog entry in the readthedocs doc build of this PR (link in github checks)
• add/modify tests,
• ready for review.

[codecov]

Codecov Report

Merging #2910 (3c54e79) into RELEASE_next_minor (6e2ddbb) will increase coverage by 0.18%.
The diff coverage is 81.25%.

@@                  Coverage Diff                   @@
##           RELEASE_next_minor    #2910      +/-   ##
======================================================
+ Coverage               79.37%   79.56%   +0.18%     
======================================================
  Files                     208      209       +1     
  Lines                   31939    32270     +331     
  Branches                 7194     7242      +48     
======================================================
+ Hits                    25353    25677     +324     
- Misses                   4837     4839       +2     
- Partials                 1749     1754       +5     

Impacted Files	Coverage Δ
hyperspy/io.py	86.45% <ø> (+0.69%)	⬆️
hyperspy/io_plugins/bruker.py	88.04% <81.25%> (-0.80%)	⬇️
hyperspy/io_plugins/__init__.py	71.42% <0.00%> (ø)
hyperspy/io_plugins/tvips.py	99.07% <0.00%> (ø)

---

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 6e2ddbb...3c54e79. Read the comment docs.

[sem-geologist]

looks that fixing the behaviour of cutoff_at_kV makes it behaves similarly to jeol reader where None does not do any hideous auto cutting, I think I should also now address and change docstring of common reader. I see it also has downsampling. I think there could be separate PR to look through all of readers and find same keywords to update. I think this PR is complete.

[ericpre]

Looks good, I have left some questions/suggestions.

[sem-geologist]

@ericpre , what do you think about the changes made? ("auto" to "last_non-zero"; "lowest_constraint" to "em_hv")

[ericpre]

Thanks, this looks good to me!
Would it be possible to clean the git history by squashing some of the commits? For example, in this PR, there are some iterative changes and for our future self, typically when browsing the git history, it is convenient not to have commits which do/undo changes.

[sem-geologist]

squashed last 12. Is it enough? I was unfortunately mobile and I don't know how on github I could edit few files and commit a bunch. Thus such an amount of commits...