Philippe 10/04/2018: lfi2cdf: take into account previous commits on dimension variables + bug correction on NKMAX

* treat more correctly datetime variables from LFI files
* pass HPROGRAM_ORIG at closing and not opening of file

* improvements
* split option works again

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

Philippe 23/02/2018: lfi2cdf: add CPROGRAM + some initialisations to later use file and fields structures from MNH

Philippe 22/02/2018: lfi2cdf: works with more situations (not yet perfect) + nc4 files have a .nc suffix

Philippe 15/11/2016: suffix for netCDF files is now .nc (instead of .nc4). This is necessary to respect CF convention.

(cherry picked from commit b5702ae3)

Philippe 24/112016: lfi2cdf: bug correction: size of buffer was too small when doing cdf2lfi (length of comment was not taken into account)

Philippe 24/11/2016: lfi2cdf: nc4 files have again a .nc4 extension to be consistent with MESONH and the scripts

* return is not allowed in main program => stop
* number of variables to be read and written was not computed correctly
* name of the outputfile is now correct
* treat correctly the last variable given with -v option
* correct bugs if LOWMEM is set (variables with length=1, reading of TEXT variables)
* correct types when LFI_INT is 8