refactor: rewrite simulation stages, improve optical config spectral considerations

[tlambert03]

this generally rewords the simulation pipeline. I think the biggest conceptual shift is probably this:

• there is no longer an illumination_flux method (with a full zyx extent); instead there is a new filtered_emission_rates() method that returns a non-spatial (C, F, W) array, representing the full spectral emission profile of each channel/fluorophore pair (without binning), in units of photons/sec.
• emission_flux still exists as a convenience method, but isn't used internally by the simulation anymore
• instead, the modality.render() method now takes in the ground_truth ((F, Z, Y, X)) and the filtered_emission_rates ((C, F, W)). it is then responsible for binning of the emission (if desired), which is only necessary for "hyperaccurate" simulations that want to represent a single channel as a superposition of many slightly different PSFs. The default is simply to bin all the emission photons (per channel/fluor) into an average wavelength and use a single PSF, but that setting is controlled by settings.spectral_bins_per_emission_channel. One advantage here is that all channels don't need to "agree" on the binning edges or strategy: the binning is an interim step that happens only within a single channel/fluorophore.
• there is a new method optical_image_per_fluor that represents the fullest expansion of dimensions in the pipeline (C, F, Z, Y, X), all it does is return modality.render; and optical_image is simply the output of that method summed over fluorophores: (C, Z, Y, X)

the rest is pretty much the same.

New tutorial added here, which will be in the docs:

https://github.com/tlambert03/microsim/blob/bdc748878df587691809d951bbe8a654fa323463/docs/tutorial.ipynb

[codecov]

Codecov Report

Attention: Patch coverage is 86.22642% with 73 lines in your changes missing coverage. Please review.

Project coverage is 84.84%. Comparing base (2c05362) to head (a9a376b).
Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
src/microsim/schema/optical_config/config.py	84.37%	20 Missing ⚠️
src/microsim/schema/optical_config/lib.py	31.57%	13 Missing ⚠️
src/microsim/schema/simulation.py	90.90%	12 Missing ⚠️
src/microsim/schema/modality/_simple.py	86.95%	9 Missing ⚠️
src/microsim/schema/spectrum.py	81.57%	7 Missing ⚠️
src/microsim/fpbase.py	62.50%	3 Missing ⚠️
src/microsim/psf.py	83.33%	3 Missing ⚠️
src/microsim/schema/sample/fluorophore.py	87.50%	3 Missing ⚠️
src/microsim/schema/sample/sample.py	90.90%	2 Missing ⚠️
src/microsim/util.py	97.61%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #74      +/-   ##
==========================================
+ Coverage   84.19%   84.84%   +0.64%     
==========================================
  Files          49       47       -2     
  Lines        2892     3094     +202     
==========================================
+ Hits         2435     2625     +190     
- Misses        457      469      +12     

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

[tlambert03]

@federico-carrara, if you have time/interest, you may wish to take a peek at this. I don't expect you to review the entire thing, but if you have any questions/concerns about your particular use case, let me know

[federico-carrara]

Hi @tlambert03!
Thank you for pointing me to this PR.
I am definitely up for having a look at it over the week as I also plan to get back to work on data simulation in a few days :)
Have a nice day!!