Pysyft: Multiple workers with same id issues

[LaRiffle]

Describe the bug
When multiple workers with the same id are declared, we observe some weird behaviour.
Could be just raise an Error when this happen? Alternatively, handle this in a consistent way.

To Reproduce

import torch
import syft as sy
hook = sy.TorchHook(torch)

worker = sy.VirtualWorker(hook, id="worker")
worker.clear_objects()
ptr = torch.tensor([1,2,3]).send(worker)
print(worker._objects) # >>> Has 1 object!

worker = sy.VirtualWorker(hook, id="worker")
worker.clear_objects()
ptr = torch.tensor([1,2,3]).send(worker)
print(worker._objects) # >>> Empty!

[amit-rastogi]

@LaRiffle I would like to take this up.

[LaRiffle]

Sure!