Changeset 5976 for branches

DEV: tested for 50 years on a single pixel with all PFTs and tested for 15 years over Europe with a land cover map. This commit contains extended comments in sechiba in sechiba_hydraul arch that refelect our findings of better understanding why GPP can become zero under severe water stress. This code also contains the implementation of a sugar loading as a possible solution for #476 of the trunk. That solution required giving up on the current calculation of gtemp in growth_fun_all which resulted in opening ticket #561. It needs to be tested whether the sugar loading variables are properly cleaned and moved when a PFT dies