Substitute pow2bin allocator with a dlmalloc based allocator (#71)

* Add dlmalloc_strat allocator strategy
 - Use mspace variant to ease encapsulation
 - Make pow2bins and dlmalloc cmake selectable
* Add unit tester for dlmalloc, rework single_heap, pow2bins unit testers
accordingly
 - add dlmalloc get_used/get_avail, and have all strats allocators also have a get_used
 - Rework memallocator unit tests: bin size is per strat, alignment is verified in singleheap
* bugfix: dlmalloc exposed that the pingpong test would write past end of
allocation with -w 32
* iostream leakage/mixed usage of cerr and fprintf(stderr

---------

Signed-off-by: Aurelien Bouteiller <aurelien.bouteiller@amd.com>

src/backend_bc.cpp

@@ -93,7 +93,7 @@ void Backend::init_mpi_once(MPI_Comm comm) {
   if (!init_done) {
     NET_CHECK(MPI_Init_thread(0, 0, MPI_THREAD_MULTIPLE, &provided));
     if (provided != MPI_THREAD_MULTIPLE) {
-      std::cerr << "MPI_THREAD_MULTIPLE support disabled.\n";
+      fprintf(stderr, "MPI_THREAD_MULTIPLE support disabled.\n");
     }
   }
   if (comm == MPI_COMM_NULL) comm = MPI_COMM_WORLD;