Merge of branch dev_r5204_CNRS_PISCES_dcy into the trunk.

[cetlod]

Summary of the branch :

• Diurnal cycle in PISCES
  • computation of the daily mean solar radiation in TOP and anymore in OPA
  • computation of 2 PAR in PISCES
• New temporal leapfrog scheme in PISCES
• Biogeochemistry in sea-ice : 2015/dev_r5171_CNRS_LIM3_seaicebgc
• Some improvements of passive tracers conservation : add a surface boundary tracer content correction in time-stepping when using vvl
• Possibility to compute a first guess of river runoff depth from a max value of runoff clim over a global domain and a depth max over which runoffs are spread
• Hard coded indices from ORCA1 extended grid
• specific output grid for zonal mean

[hadcv]

There are a couple of changes that cause SETTE to fail (for me) at the moment:

• ORCA2_LIM_OBS is added in https://forge.ipsl.jussieu.fr/nemo/changeset/5385/trunk/NEMOGCM/CONFIG/cfg.txt but is not an existing configuration in the CONFIG directory
• namtrc_dmp has no closing character in https://forge.ipsl.jussieu.fr/nemo/changeset/5385/trunk/NEMOGCM/CONFIG/SHARED/namelist_top_ref

[hadcv]

I have committed the above fixes at [5416]; SETTE now passes successfully for me

[clevy]

Suppressing tickets opened on the trunk since more than 2 years... cleaning started!