Minor fixes: n/k database, updating examples, bugs in Cauchy model and material parameters

[phoebe-p]

Few minor fixes:

• Update which version of the refractiveindex.info database gets downloaded by default (was a version from 2018, now the most recent version). This changes the "pageid" which has to be used in some of the examples. In the example that is explicity about this I just updated the numbers and made a note that depending on the database version these may be different; in the 4J cell optimization example it is now fixed so that it will find the right entry regardless of database version. These examples have also been updated in the documentation
• The Cauchy model to calculate the refractive index had a bug (final term was C*wavelength^4 instead of C/wavelength^4. This is fixed.
• I am not sure if this was intentional, but for Si and Ge there was an entry in their parameter files for relative_dielectric_constant rather than relative_permittivity. relative_dielectric_constant is never actually used by Solcore (unlike relative_permittivity) and as far as I understand it is the same thing anyway. This means that if you used Si and Ge you had to set this parameter manually, which seems unnecessary given that it is well-known, so I have changed these entries to relative_permittivity.
• Similar to the issue fixed in this commit, now also fixed in the PDD solver; if you did not explicity set an incident spectrum, but did specify the wavelengths, then your user-defined wavelength would get overwritten when calculating the spectrum in short_circuit_pdd

[codecov]

Codecov Report

Merging #218 (80ff2b1) into develop (b8b8b44) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff            @@
##           develop     #218   +/-   ##
========================================
  Coverage    45.85%   45.86%           
========================================
  Files           84       84           
  Lines         9100     9101    +1     
========================================
+ Hits          4173     4174    +1     
  Misses        4927     4927           

Impacted Files	Coverage Δ
solcore/absorption_calculator/nk_db.py	17.74% <ø> (ø)
solcore/solar_cell_solver.py	66.46% <ø> (ø)
...bsorption_calculator/dielectric_constant_models.py	75.32% <100.00%> (+0.16%)	⬆️
...erial_data/refractiveindex_info_DB/dboperations.py	14.13% <100.00%> (ø)
...poisson_drift_diffusion/DriftDiffusionUtilities.py	86.47% <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 b8b8b44...80ff2b1. Read the comment docs.

[dalonsoa]

All good! Well spotted errors and updates.

[dalonsoa]

Considering the severity of the error, it suggests this model has not been used that often or someone should have spotted the mistake long ago.

[phoebe-p]

Yes, I tried to actually use it and it was immediately obvious that something was wrong, but that was the first time I (and maybe anyone) had actually tried to use it...

[dalonsoa]

I wish there were a more programatic way of checking the database data and fetching the last version. Hardcoding it looks a bit hacky.

[phoebe-p]

There probably is, I just haven't looked into it! I will leave it like this now but will have a look to see if there is a more elegant way I can do it.