This is to allow front-end to properly handle ID values.
This change was done to keep consistent with what we were doing
before. Alternative version would be to have a sepparate struct
on the rest package.
Lapsus in component ident field caused too much logic to be
removed from the generated code.
This caused some _happy accidents_ when matching up to existing
resources.
* Fix invalid RBAC rule unmarshling where rules under resources
incorrectly set resource references.
* Make reference assumptions not so aggressive where they will
only apply when missing.
* Add missing skipIf evaluation checks.
Matchups were not scoped which caused them to get matched
to any matched ident, regardles of scope.
This caused NS duplication to "move" original resources to the
cloned one instead of copying them.
Passing dialect options lower so the supporting functions can prepare
proper expressions.
Might be better off defining a new nuance for this, or researching
more if I can trick goqu to do it correctly.
Updated parse sorting method to support coalesce sorting with multiple columns, by extending the current sortExpr with modifier and multiple columns, IE. `coalesce` as modifier and given arguments will be multiple columns for the sortExpr.
Also, updated cursor to support modifier and multiple column support along with rdbms.go.tpl for the same.
