Kf/fix pytorch warnings

[KasperFyhn]

Two warnings are triggered in the Multi2OIE code:

/home/au479461/PycharmProjects/conspiracies/src/conspiracies/docprocessing/relationextraction/multi2oie/other/bio.py:209: UserWarning: Creating a tensor from a list of numpy.ndarrays is extremely slow. Please consider converting the list to a single numpy.ndarray with numpy.array() before converting to a tensor. (Triggered internally at  ../torch/csrc/utils/tensor_new.cpp:210.)
  return [torch.Tensor(pred_tags) for pred_tags in total_pred_tags]
/home/au479461/PycharmProjects/conspiracies/src/conspiracies/docprocessing/relationextraction/multi2oie/other/bio.py:38: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  return torch.tensor(res, dtype=torch.bool, device=tensor.device)

This PR addresses those two warnings.

1. vstacks numpy arrays instead of lists in a loop and creates a tensor from the resulting array, as the warning suggests.
2. Uses tensor.bool() to convert the original tensor straight into a boolean tensor that gets modified based on the contents of the original tensor, thereby skipping one tensor construction and avoiding the warning from the constructor.

[KennethEnevoldsen]

Love these changes Kasper!