Ensure casting to typmod -1 generates a RelabelType.

Fix the code changed by commit 5c056b0c2 so that we always generate
RelabelType, not something else, for a cast to unspecified typmod.
Otherwise planner optimizations might not happen.

It appears we missed this point because the previous experiments were
done on type numeric: the parser undesirably generates a call on the
numeric() length-coercion function, but then numeric_support()
optimizes that down to a RelabelType, so that everything seems fine.
It misbehaves for types that have a non-optimized length coercion
function, such as bpchar.

Per report from John Naylor.  Back-patch to all supported branches,
as the previous patch eventually was.  Unfortunately, that no longer
includes 9.6 ... we really shouldn't put this type of change into a
nearly-EOL branch.

Discussion: https://postgr.es/m/CAFBsxsEfbFHEkouc+FSj+3K1sHipLPbEC67L0SAe-9-da8QtYg@mail.gmail.com

src/test/regress/sql/expressions.sql

@@ -84,6 +84,22 @@ create view numeric_view as
 
 explain (verbose, costs off) select * from numeric_view;
 
+-- bpchar, lacking planner support for its length coercion function,
+-- could behave differently
+
+create table bpchar_tbl (f1 character(16) unique, f2 bpchar);
+
+create view bpchar_view as
+  select
+    f1, f1::character(14) as f114, f1::bpchar as f1n,
+    f2, f2::character(14) as f214, f2::bpchar as f2n
+  from bpchar_tbl;
+
+\d+ bpchar_view
+
+explain (verbose, costs off) select * from bpchar_view
+  where f1::bpchar = 'foo';
+
 rollback;