The code was mistakenly calling gimp_ui_init to init gegl,
and then gtk_init was aborting in headless gimp-console.
Now require GimpPlugin class to init gegl/babl.
See next commit.
Call gimp_ui_init only when run mode INTERACTIVE.
Call gimp_ui_init during query phase as well as run phase.
Maybe gimp_main() should do this.
Since space invasion from GimpRGB to GeglColor,
plugins that declare defaults for formal args of type GeglColor
need babl and gegl at plugin registration time.
Fixes errors introduced by my prior commit c0ca0b74
1. SF is storing colors in GimpRGB, not GeglColor.
Storing using wrong pointer type seemed to work, but would not build on newer compilers.
2. spurious error message "CRITICAL....Two different plugins tried to register..."
Due to calling gimp_ui_init too early, in the class instance init.
Since PDB now traffics in GeglColor instead of GimpRGB.
Change GimpParamSpecRGB to GeglParamSpecColor when declaring args to procedure.
Declared default is temporarily "black".
Change conversion of gvalue to scheme representation, now from GeglColor.
Incidental change to interpreter to init gegl early,
since now needed during registration phase.
Why:
1) users can install .scm scripts to plug-ins dir
2) Crashing scripts do not crash extension-script-fu
Scripts (.scm files) have a shebang and are executable
and in a same-named subdir of plugin dir.
Interpreter/scripts create PDB procs of type PLUGIN unlike extension-script-fu
which creates PDB procs of type TEMPORARY, owned by extension-script-fu.
Unlike other interpreters, the interpreter is-a plugin outright,
not by virtue of the script subclassing GimpPlugin and using GI.
More details in /plug-ins/script-fu/interpreter/README
