Objective function evaluation on GPU with minimal PCIe transfers (#2935)

* Added GPU objective function and no-copy interface.

- xgboost::HostDeviceVector<T> syncs automatically between host and device
- no-copy interfaces have been added
- default implementations just sync the data to host
  and call the implementations with std::vector
- GPU objective function, predictor, histogram updater process data
  directly on GPU

src/common/device_helpers.cuh

@@ -484,6 +484,13 @@ class bulk_allocator {
   }
 
  public:
+  bulk_allocator() {}
+  // prevent accidental copying, moving or assignment of this object
+  bulk_allocator(const bulk_allocator<MemoryT>&) = delete;
+  bulk_allocator(bulk_allocator<MemoryT>&&) = delete;
+  void operator=(const bulk_allocator<MemoryT>&) = delete;
+  void operator=(bulk_allocator<MemoryT>&&) = delete;
+  
   ~bulk_allocator() {
     for (size_t i = 0; i < d_ptr.size(); i++) {
       if (!(d_ptr[i] == nullptr)) {