Fix beam normalization

[paganol]

This pull request updates the polarization normalization in the Gaussian beam synthesis routine to match the convention used in the gauss_bl() function. As a result, the expected values in the corresponding unit tests have been updated to reflect the new normalization.

Beam synthesis normalization changes:

• Added Challinor spin-2 normalization factor (exp(2*sigma^2)) for polarization modes in gauss_beam_to_alm to match the normalization used in gauss_bl().

Test updates for new normalization:

• Updated expected values in test_gaussbeam_circular in test/test_beam_synthesis.py to reflect the new polarization normalization.
• Updated expected values in test_gaussbeam_elliptical in test/test_beam_synthesis.py for consistency with the new normalization.

[github-actions]

Coverage report

Click to see where and how coverage changed

File	Statements	Missing	Coverage	Coverage (new stmts)	Lines missing
litebird_sim
beam_synthesis.py
Project Total

This report was generated by python-coverage-comment-action

[paganol]

Now the difference input VS convolved goes down to $10^{-19}$ K in polarization if we set epsilon=1e-13 in BeamConvolutionParameters. With the default, i.e. epsilon=1e-5, we reach $10^{-11}$ K. updated version of the notebook
Here an updated version of the notebook

[paganol]

@ziotom78, probably this fix could be added to the v0.16.x branch