Decrease number of concurrent threads in concurrent hashmap tests if run under
drd. It fix timeout issue during travis builds.
Varialbe "On_drd" is exposed by unittest library.

Add tests for all kinds of insert, erase and find, as defined in API.
 * Abstraction of class ConcurrentHashmapTestPrimitives is added to achieve
tests code as self documented scenarios.
 * Prints are not visible during ctest execution, but are helpful when
test binary is run manually.

string: modifiers

Make valgrind build separate travis step

It's done for build time optimization.

* Put builds as separate functions in run-build.sh
* Explicitly list, which builds should be run according to build type
* Add valgrind build type
* Add valgrind build as separate travis step

Add concurrent_hash_map_pmreorder_break_insert test

Add simplekv example

Add concurrent_hash_map_pmreorder_erase test

Add flags to disable warning C4996

Fix warnings C4800: 'int': forcing value to bool

Add concurrent_hash_map_pmreorder_insert_erase_mt test

Add flags:
 -D_SCL_SECURE_NO_WARNINGS to disable checking iterators
 -D_CRT_SECURE_NO_WARNINGS to disable deprecation of getenv and strcpy functions

It disables 156 such warnings.

Ref: #409

Correct small misspell in pman's Game Over message

It fixes 503 warnings #C4800.

Ref: #409

and rebuilds rest on application restart.

Decrease Visual Studio warning level to W3

appveyor: msbuild: set verbosity to minimal

... to reduce the Appveyor build time.

... to reduce the Appveyor build time.

appveyor: set PMEM_IS_PMEM_FORCE in tests

PMEM_IS_PMEM_FORCE is not set in tests
and Appveyor build is timed out.

Appveyor: switch to Visual Studio 2017

This change is required because of the following issue:
https://help.appveyor.com/discussions/problems/24684-problem-with-vcpkg-wrong-version-on-appveyor
https://github.com/appveyor/ci/issues/3052
which cannot be fixed directly now.

Concurrent hashmap: removal of atomic allocation support and minor refactoring

Ref: https://github.com/pmem/libpmemobj-cpp/issues/356

Atomic allocator has limited use case since pmemobj does not allow
for nesting make_persistent_atomic calls.