Previously we'd only check whether the hardening options succeeded
at the compile step. Now we'll try to link with them too, and tell
the user in advance if something seems likely to go wrong.

Closes ticket 18895.

I'm doing this to simplify crypto_digest_smartlist_prefix, and make
it better covered by our tests.

Apparently, we detect and reject them correctly. Aren't you glad?

Fix for bug 18956.

Also, resolve a bug in test_ntor_cl.c

Had conflicts related to other correct_tm bugs in 0.2.6.  Added wday
for another case.

The tm_wday field had been left uninitialized, which was causing
some assertions to fail on Windows unit tests.

Fixes bug 18977.

This is a whitespace only, cosmetic fix.

There is still some inconsistency between lists, but less
inconsistency inside individual lists.

There were some conflicts here, and some breakage to fix concerning
library link order in newer targets.

Fixes bug 18901.

This way it gets the ctime options.

This type saved a tiny amount of allocation, but not enough to be
worth keeping.

(This is in preparation for moving choose_array_element_by_weight)

We know there are overflows in curve25519-donna-c32, so we'll have
to have that one be fwrapv.

Only apply the asan, ubsan, and trapv options to the code that does
not need to run in constant time.  Those options introduce branches
to the code they instrument.

(These introduced branches should never actually be taken, so it
might _still_ be constant time after all, but branch predictors are
complicated enough that I'm not really confident here. Let's aim for
safety.)

Closes 17983.

Closes 19041.