improving docstring (and definition?!) of mean free path in diffusion kinetics

[bhiogade]

No description provided.

[slayoo]

lambdaD is the vapour molecular mean free path, however it seems that we might have a bug in the definition, at least as compared with libcloudph++:

return real_t(2) * (D_0<real_t>() / si::metres_per_second) / (
          real_t(sqrt(
            real_t(2) * real_t((R_v<real_t>() / si::joules * si::kilograms) * T) 
            ))
          );

(https://github.com/igfuw/libcloudphxx/blob/0ca09ab6ba9d8b6a6096ef4dbc516ee22c7d551e/include/libcloudph%2B%2B/common/mean_free_path.hpp#L25-L29)

vs. in PySDM:

PySDM/PySDM/physics/diffusion_kinetics/fuchs_sutugin.py

Line 15 in b476bb0

return D / np.sqrt(2 * const.Rv * T)

and

PySDM/PySDM/physics/diffusion_kinetics/jensen_and_nugent_2017.py

Line 14 in b476bb0

return D / np.sqrt(2 * const.Rd * T)

and

PySDM/PySDM/physics/diffusion_kinetics/pruppacher_and_klett_2005.py

Line 15 in b476bb0

return D / np.sqrt(2 * const.Rv * T)

(BTW, the same definition should not be repeated in the above three places!)

[github-actions]

Stale pull request message