Print fully qualified type name in case of type mismatch

[jhalakpatel]

Passing a Torch tensor to Tripy operations produces a cryptic error message since both Tripy and Torch use the same qualified name i.e. "Tensor" for their corresponding Tensor class i.e. torch.Tensor and tripy.frontend.tensor.Tensor.

    def test_tensor_mismatch_tripy(self):
        a = torch.arange(1)
        b = torch.arange(1)
        print(tp.equal(a, b))

Error:

E For parameter: 'a', expected an instance of type: 'Tensor' but got argument of type: 'Tensor'.

Fix is to simplify prepend obj.__qualname__ with obj.__module__ to generate a fully qualified name.

E For parameter: 'a', expected an instance of type: 'tripy.frontend.tensor.Tensor' but got argument of type: 'torch.Tensor'.

[jhalakpatel]

@pranavm-nvidia will submit a PR that fixes the type mismatch issue and some simplifications. Closing this PR in favor of it.