Force correct self.technosphere_matrix format for solver

[marc-vdm]

What
self.technosphere_matrix is now forced to csr when using PYPARDISO or otherwise forced to csc

Why

1. This forces the correct matrix format for the solver, e.g. when user is over-writing their own technosphere, the correct format is still ensured
2. For scipy users, this shouldn't change anything
3. For PyPardiso users, this should speed up calculations ~10% on repeated solves, as BW would previously force the format to csc and then pypardiso would force it back to csr

• Fix Factorizing matrices with pypardiso is not a no-op #98

See here for further discussion: haasad/PyPardiso#75 (comment)

[codecov]

Codecov Report

Attention: Patch coverage is 47.05882% with 9 lines in your changes missing coverage. Please review.

Project coverage is 81.44%. Comparing base (2417a1b) to head (a89ede1).
Report is 1 commits behind head on main.

Files	Patch %	Lines
bw2calc/lca_base.py	41.66%	7 Missing ⚠️
bw2calc/multi_lca.py	60.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #101      +/-   ##
==========================================
- Coverage   83.65%   81.44%   -2.22%     
==========================================
  Files          14       14              
  Lines         832      846      +14     
==========================================
- Hits          696      689       -7     
- Misses        136      157      +21     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[marc-vdm]

Maintainers, please advise on failing tests.

1. The tests run fine for me locally (linux, py3.12), so I can't reproduce this as MC apparently works here
2. I have no clue what Codecov wants from me.

[cmutel]

I am going to merge and fix tests afterwards