more flexible mechanism for having different equations between the static and the dynamic model

Currently, if one wants to tell Dynare that an equation has a different form between the static and the dynamic model (for example when the modeler wants to use his knowledge about the steady state to simplify the steady state computation), then the possibilities are:

• write a steady state file (heavy)
• write a steady_state_model block (one has to rewrite all the equations in that block)
• use a dirty hack with the macroprocessor and run Dynare two times

The steady_state_model is good for small models. But for big models where you want to change only a few equations, only the third option of this list is reasonable, and it is dirty.

The proposal is to introduce two equation tags which will have a special meaning to the preprocessor:

• [static]: this equation belongs to the static model, but not to the dynamic model
• [dynamic]: the opposite

This would offer a very flexible way to change only a few equations during the steady state computations.

This is a request by @FerhatMihoubi for GPM and possibly GIMF.