fclen > 2**31 seconds

[antonl321]

This PR introduces code changes that allow IFS forecast length to go beyond 2^31-1 seconds.
This required to promote a few variable of 64bit integers and interface changes for two subroutines.

Also I think that I found bug, see below.

Structural or significant code changes:

• incdate.F90 was transform into a module yowincdate.F90 to provide for the time shift kinds with a generic interface
• difdate subroutine returns a 64bit integer.
• in wavemdl.F90 the block of code that does incremental change of the end date (CDATEF) was replaced with one call to incdate for it can use 64 integers.

Possible bug
it seems that a division by 3600 was missed in
src/ecwam/outmdldcp.F90 line ~ 127 for the variable IFCST
see also the documentation of the subroutine that uses the variable IFCST, src/ecwam/wgribencode:F90
and e.g.
src/ecwam/outspec.F90
in which the division by 3600 is done

[antonl321]

I have agregates incdate and difdate in one module (yowdate_utils.F90) with generic interfaces to allow 64 bits integer argument. In ifs-source I have introduced yowdate_utils acrorss wam subdirectory.

[antonl321]

unfortunately some of the commets were made in the rejected PR #132
I think we have to start afresh with this one.

[jrbidlot]

I assume that this work is connected to DestinE running.
The change in ecwam make sense to me.
You mentioned that you have introduced yowdate_utils across wam subdirectory.
Will I need to review the ifs-source branch?

[antonl321]

Yes, I have adapted ifs-source consistently with the work done in ecwam.
The bulk of the changes related to ecwam are in this commit
https://github.com/antonl321/ifs-source/commit/1fd239554f271138d53f6d3ccfa8cbc7152ace04

I don't have yet a PR ready for ifs-soure.

[jrbidlot]

Hi Julian,
when you have a PR, make sure to also include James Steer in the review. A large part of the change in ifs-source wam will affect the wave data assimilation, which is now James's responsibility