[m-rev.] for review: disable_warnings scope

[Zoltan Somogyi]

This is the scope several of us discussed on friday.

For review by Paul. I would particularly like to know why warnings
about recursive calls that are not TAIL recursive are generated in
two separate places, and whether there is any particular reason why
we can't delete the version in ml_tailcalls.m, and make mark_tail_calls.m
do the job in all grades. I also added some XXXs that I would like his
opinion on.

I would welcome reviews of the new construct's syntax, and its
documentation, by others as well.

Zoltan.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Log.dw
Type: application/octet-stream
Size: 3611 bytes
Desc: not available
URL: <http://lists.mercurylang.org/archives/reviews/attachments/20170109/27534064/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: DIFF.dw
Type: application/octet-stream
Size: 70337 bytes
Desc: not available
URL: <http://lists.mercurylang.org/archives/reviews/attachments/20170109/27534064/attachment-0003.obj>