Fix dask.array.random 2023.2.1

[ericpre]

Dask 2023.2.1 refactor the code to generate random numbers and now follows the syntax introduced in numpy 1.17 - see #3049 and https://numpy.org/doc/stable/reference/random/new-or-different.html#new-or-different for more details.

Progress of the PR

• Deprecate use of numpy.random.RandomState and start to use np.random.default_rng in test suite instead,,
• Bump scipy minimum requirement to 1.4.0 to support np.random.default_rng,
• update test depending on numpy.random.RandomState,
• [n/a] update docstring (if appropriate),
• [n/a] update user guide (if appropriate),
• 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 tests,
• ready for review.

[codecov]

Codecov Report

Patch coverage: 90.00% and project coverage change: -0.01 ⚠️

Comparison is base (cf21c21) 81.00% compared to head (7fcafad) 81.00%.

Additional details and impacted files

@@                  Coverage Diff                   @@
##           RELEASE_next_patch    #3103      +/-   ##
======================================================
- Coverage               81.00%   81.00%   -0.01%     
======================================================
  Files                     209      209              
  Lines                   32753    32770      +17     
  Branches                 7237     7242       +5     
======================================================
+ Hits                    26533    26545      +12     
- Misses                   4466     4470       +4     
- Partials                 1754     1755       +1     

Impacted Files	Coverage Δ
hyperspy/signal.py	77.08% <ø> (ø)
hyperspy/signal_tools.py	73.83% <66.66%> (+0.03%)	⬆️
hyperspy/misc/math_tools.py	79.22% <95.65%> (-3.04%)	⬇️
hyperspy/samfire.py	65.54% <100.00%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[ericpre]

@CSSFrancis, @jlaehne, @francisco-dlp, can one of you review this PR?

[CSSFrancis]

I have a bit of free time and can do it now!

[CSSFrancis]

This seems like all pretty straight forward changes. The only thing that I thought might be worth doing extra is adding a comment where you change a couple of the tests incase someone is testing with an older version of numpy and those tests start to fail. It might give some idea of why that is the case.