[m-rev.] for review: better error message for bad type names

[Zoltan Somogyi]

For review by anyone, but I would like Mark to have a look
at the XXXs on the definition of the why_no_ho_inst_info type
in parse_type_name.m, which lists the contexts in which types
may not contain higher order inst information.

Zoltan
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Log.parse_type_name
Type: application/octet-stream
Size: 1418 bytes
Desc: not available
URL: <http://lists.mercurylang.org/archives/reviews/attachments/20160418/9a0066de/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: DIFF.parse_type_name
Type: application/octet-stream
Size: 53440 bytes
Desc: not available
URL: <http://lists.mercurylang.org/archives/reviews/attachments/20160418/9a0066de/attachment-0003.obj>