Wrong device in ttv_tensor.py bug fixed

[alfatum]

Fix for this bug:

File ".../pyCP_APR/pyCP_APR/torch_backend/ttv_sptensor.py", line 55, in ttv
    newsubs = subs[:, remdims]
RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cuda:1)

Bug can be reproduced by setting device_num != 0 in CP_APR and fitting the model. Example:

cp_apr = CP_APR(n_iters=15, random_state=42, verbose=1, method='torch', device='gpu', device_num=1)
factors = cp_apr.fit(coords=train_coords, values=train_values, rank=[1,5])