absorbed radiation is not zero if LAI is zero

[kmdeck]

Purpose

Fixes #645
We compute transmitted radiation as proportional to exp(-KLAIOmega), where Omega is O(1). The absorbed radiation is proportional to (1- exp(-KLAIOmega)).

When the sun is at the horizon, K ~ 1/eps(FT), and no radiation is transmitted (the extinction coefficient K -> infinity).

However, when LAI -> 0, we need to have exp(-KLAIOmega) -> 1, so absorbed radiation -> 0. The issue is that if LAI is O(eps(FT)), K*LAI ~O(1), and absorbed radiative fraction is O(1/e).

Solution:
Im not sure. I dont understand the behavior of K with zenith angle

To-do

deal with 1/cos(theta) when theta = pi/2
deal with exp(-K*LAI) when K ~ 1/eps(FT) (from 1/cos(theta)) and LAI ~ eps(FT)

Content

added failing test

---

• I have read and checked the items on the review checklist.

[kmdeck]

@Espeer5 in case you have any ideas, but I can implement/try and fix

[kmdeck]

Im closing this PR because for now we are clipping LAI