Improving the optical depth calculation

[cosimoNigro]

Hello,

This PR tries to improve the optical depth calculations.
It solves issue #16 though the agreement with Finke is not yet good.

Here a comparison of the optical depths for Disk

agnpy absorption starts much earlier (a factor ten in energy) than the one in Finke (2016).
Additionally agnpy absorption uses the full integration: if I try to use the simplified formula in Finke (2016) - green line (blob moving parallel to the jet) I get an absorption similarly starting at lower energies than the one of the figure sampled from the paper plot (blue line).

The absorption for the Broad Line Region and Dust Torus instead starts at larger energies than in Finke

Note though that agnpy's absorption is self-consistent. If I replace the dust torus and the BLR with a monochromatic point-like source approximating them at large distances (larger than the size of the target) I have a very good agreement (I did the same tests for the external Compton).

I don't know what am I doing wrong, any suggestion @jsitarek, @pawel21?
@jsitarek you still have that code you used to compute optical depths, might we use it for crosschecks?

[sourcery-ai]

Sourcery Code Quality Report

❌  Merging this PR will decrease code quality in the affected files by 1.56%.

Quality metrics	Before	After	Change
Complexity	0.25 ⭐	0.33 ⭐	0.08 👎
Method Length	89.49 🙂	90.93 🙂	1.44 👎
Working memory	9.90 😞	10.48 😞	0.58 👎
Quality	73.37% 🙂	71.81% 🙂	-1.56% 👎

Other metrics	Before	After	Change
Lines	1191	1517	326

Changed files	Quality Before	Quality After	Quality Change
agnpy/absorption.py	71.93% 🙂	69.92% 🙂	-2.01% 👎
agnpy/compton.py	75.90% ⭐	75.90% ⭐	0.00%
agnpy/tests/test_absorption.py		65.38% 🙂
agnpy/tests/test_compton.py	67.70% 🙂	66.86% 🙂	-0.84% 👎
agnpy/tests/test_synchrotron.py	71.51% 🙂	71.95% 🙂	0.44% 👍
agnpy/tests/utils.py	74.63% 🙂	71.43% 🙂	-3.20% 👎
docs/conf.py	68.36% 🙂	61.82% 🙂	-6.54% 👎

Here are some functions in these files that still need a tune-up:

File	Function	Complexity	Length	Working Memory	Quality	Recommendation
agnpy/tests/utils.py	make_comparison_plot	3 ⭐	243 ⛔	22 ⛔	39.30% 😞	Try splitting into smaller methods. Extract out complex expressions
agnpy/absorption.py	Absorption._opacity_disk	3 ⭐	235 ⛔	17 ⛔	43.11% 😞	Try splitting into smaller methods. Extract out complex expressions
agnpy/tests/test_synchrotron.py	TestSynchrotron.test_ssa_reference_sed	0	168 😞	35 ⛔	45.12% 😞	Try splitting into smaller methods. Extract out complex expressions
agnpy/compton.py	ExternalCompton._sed_flux_disk	0	266 ⛔	15 😞	46.64% 😞	Try splitting into smaller methods. Extract out complex expressions
agnpy/absorption.py	tau_disk_finke_2016	0	202 😞	18 ⛔	47.40% 😞	Try splitting into smaller methods. Extract out complex expressions

Legend and Explanation

The emojis denote the absolute quality of the code:

• ⭐ excellent
• 🙂 good
• 😞 poor
• ⛔ very poor

The 👍 and 👎 indicate whether the quality has improved or gotten worse with this pull request.

---

Please see our documentation here for details on how these metrics are calculated.

We are actively working on this report - lots more documentation and extra metrics to come!

Let us know what you think of it by mentioning @sourcery-ai in a comment.

[codecov]

Codecov Report

Merging #46 into master will increase coverage by 7.67%.
The diff coverage is 99.62%.

@@            Coverage Diff             @@
##           master      #46      +/-   ##
==========================================
+ Coverage   83.23%   90.90%   +7.67%     
==========================================
  Files          15       16       +1     
  Lines        1312     1462     +150     
==========================================
+ Hits         1092     1329     +237     
+ Misses        220      133      -87     

Flag	Coverage Δ
#unittests	90.90% <99.62%> (+7.67%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
agnpy/tests/utils.py	98.11% <97.91%> (-1.89%)	⬇️
agnpy/absorption.py	100.00% <100.00%> (+83.01%)	⬆️
agnpy/compton.py	86.63% <100.00%> (ø)
agnpy/tests/test_absorption.py	100.00% <100.00%> (ø)
agnpy/tests/test_compton.py	100.00% <100.00%> (ø)
agnpy/tests/test_synchrotron.py	100.00% <100.00%> (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 c073c48...4eb4fc7. Read the comment docs.

[cosimoNigro]

I am merging, let us continue the discussion with issues.