Philippe 29/03/2018: problem correction in Rules for gfortran: do not replace AINT by INT (AINT is for reals, INT for integers)
This was the origin of problems with management of XUNDEF in SURFEX
because XUNDEF can not be represented by a standard integer

+ remove temporal_lt function

Philippe 28/03/2018: replace TEMPORAL_DIST by DATETIME_DISTANCE and ADD_FORECAST_TO_DATE and SUBTRACT_TO_DATE by DATETIME_CORRECTDATE

(LTIMEDEP is only set for some test variables)

* set JPXKRK to NLFIMAXCOMMENTLENGTH
* add NSTRINGCHUNKSIZE parameter in mode_netcdf (instead of IMULT)
* typo

Philippe 22/03/2018: IO: use CFORMAT field of TFILEDATA instead of relying on LIOCDF4, LLFIOUT and LLFIREAD
This prevents some problems when mixing LFI and netCDF files (ie in cdf2lfi)
+ cleaning use of modd_io_ll for LIOCDF4, LLFIOUT and LLFIREAD (and others)

* improvements
* split option works again

Philippe 19/03/2018: IO: Z-split files: pass also the optional HFORMAT argument (necessary for LFI2CDF)

Philippe 16/03/2018: lfi2cdf: remove file suffix from hinfile and houtfile to manage more cleanly filenames

Philippe 13/02/2018: IO_GUESS_DIMIDS_NC4: manage variables with CLBTYPE/=NONE + add a return value + problems are now WARNING instead of FATAL

* write NF90_INT1 instead of NF90_INT for logicals
* use IO_WRITE_FIELD (not yet done everywhere)
* if NTYPE is unknown => fatal error
* do not try to read JPVEXT (never written)

Philippe 09/03/2018: lfi2cdf: remove redundant fields of workfield (present also in TFIELDDATA) + remove read_data_lfi