In this commit:
* Change the lit_int_unsuffixed AST node to not carry a type, since
it doesn't need one
* Don't print "(unsuffixed)" when pretty-printing unsuffixed integer
literals
* Just print "I" instead of "(integral)" for integral type variables
* Set up trans to use the information that will be gathered during
typeck to construct the appropriate constants for unsuffixed int
literals
* Add logic for handling int_ty_sets in typeck::infer
* Clean up unnecessary code in typeck::infer
* Add missing mk_ functions to middle::ty
* Add ty_var_integral to a few of the type utility functions it was
missing from in middle::ty
There was a FIXME noting that ty::enum_variants and typeck::check::
check_enum_variants both call eval_const_expr. I tried refactoring the
code so that check_enum_variants does all the work and enum_variants
just looks up cached results, but this turned out not to be easy because
several ty functions call enum_variants and it might get invoked on an
enum before that enum item has been typechecked. Instead, I just made
check_enum_variants update the enum_var_cache so that enum_variants
will never call eval_const_expr twice on the same const.
Issue 2526 showed a test case where a library exported only a type
that was a synonym for a class. Because the class's destructor wasn't
getting marked as reachable, its linkage was wrongly getting set to
"internal". The solution is for reachability to traverse types.
Closes#2526.
And replace trans::common::block_parent with option<block>. To handle
the recursive self-reference in the block_ class, I had to add a
newtype-like enum "block" which is equivalent to @block_ -- which due
to an interaction with borrowck, resulted in having to change a few
functions in trans::base to take their block argument in ++ mode,
irritatingly enough (but not that irritatingly, since we're supposed to
get rid of modes).
