Fixes and Cleanup

[austin-beer]

• Fixed build failures on Windows. The timespec struct is not supported by older versions of Visual Studio. I changed the internal representation inside of the *_timespec_timepoint classes to a boost::intmax_t representing the number of nanoseconds since the epoch.
• Fixed some functions that wouldn't execute at all if they were provided a negative time duration or an absolute time that was in the past. From what I understand, they should instead execute once and then return immediately.
• Moved pthread/timespec.hpp to detail/timespec.hpp.
• Deleted detail/internal_clock.hpp and moved the seven relevant lines into detail/timespec.hpp. This keeps all of the internal clock declarations in one place.
• Renamed thread_detail::internal_clock_t to detail::internal_chrono_clock to be consistent with and yet clearly differentiated from detail::internal_timespec_clock.
• Removed "using namespace chrono" to eliminate ambiguious namespace resolution when referencing detail::internal_chrono_clock.
• Re-enabled a few tests on Windows that had previously been disabled. I want to see whether or not they still need to be disabled.

I'm hoping this PR will fix the AppVeyor failures that the last PR introduced. I have been building with Visual Studio 2015 but I switched over to Visual Studio 2010 to better catch issues like this in the future.

[viboes]

Ok, I will add an old version of msvc into AppVeyor to check this kind of backward compatibility issues. Maybe you can do it on your branch.

[viboes]

There are a lot of errors in Windows and also on MacOs :(

[austin-beer]

Yep, I'm working on those as we speak.

[viboes]

The issue from

libs/thread/test/sync/mutual_exclusion/sync_pq/pq_single_thread_pass.cpp(58): test 'diff < milliseconds(600) && diff > milliseconds(500)' failed in function 'void test_pull_for()'

is not your fault.

The implementation/interface of sync_priority_queue is wrong. I'll try to fix it.

Hrr, there are too much wrong time related interfaces in Boost.Thread.

[viboes]

I don't remember why I dded such time-related functions as the C++ proposal it is based on (http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2017/p0260r1.html) don't provides such functions.

[viboes]

If you believe your PR is ready to merge I'll do.

[viboes]

Could you give a last try with

  BOOST_TEST(diff < milliseconds(650) && diff > milliseconds(500));

I suspect that now that we could be pooling (at least in MacOs) waiting for 500ms could be very easily wait for more than 600ms.

[viboes]

As you see I've already merged your PR.

[austin-beer]

Thanks. I've been out of the office all weekend so I wasn't able to take care of this. Thanks for taking a look at the pq_single_thread_pass issues.