[LDistance] Fix OpenMP segfault on Windows

Weird interaction between the LDistanceMatrix parallel loop (with the
dynamic scheduling) and the LDistance reduction loop that should be
sequential (threadNumber_ == 1).

core/base/lDistanceMatrix/LDistanceMatrix.h

@@ -53,8 +53,7 @@ int ttk::LDistanceMatrix::execute(std::vector<std::vector<double>> &output,
 
   // compute matrix upper triangle
 #ifdef TTK_ENABLE_OPENMP
-#pragma omp parallel for num_threads(this->threadNumber_) schedule(dynamic) \
-  firstprivate(worker)
+#pragma omp parallel for num_threads(this->threadNumber_) firstprivate(worker)
 #endif // TTK_ENABLE_OPENMP
   for(size_t i = 0; i < nInputs; ++i) {
     for(size_t j = i + 1; j < nInputs; ++j) {