Stop using an unnecessary scalar_to_tensor(..., device) call.

[gchanan]

Stack from ghstack:

• Stop moving scalars to GPU for one computation in leaky_rrelu_backward. #50115 Stop moving scalars to GPU for one computation in leaky_rrelu_backward.
• Stop using an unnecessary scalar_to_tensor(..., device) call. #50114 Stop using an unnecessary scalar_to_tensor(..., device) call.
• Move scalar_to_tensor_default_dtype out of ScalarOps.h because it's only useful for torch.where. #50111 Move scalar_to_tensor_default_dtype out of ScalarOps.h because it's only useful for torch.where.
• Stop using c10::scalar_to_tensor in float_power. #50105 Stop using c10::scalar_to_tensor in float_power.

In this case, the function only dispatches on cpu anyway.

Differential Revision: D25790155

[facebook-github-bot]

💊 CI failures summary and remediations

As of commit da49573 (more details on the Dr. CI page):

---

• 1/1 failures possibly* introduced in this PR
  • 1/1 non-CircleCI failure(s)

---

This comment was automatically generated by Dr. CI (expand for details).

Follow this link to opt-out of these comments for your Pull Requests.

Please report bugs/suggestions to the (internal) Dr. CI Users group.

This comment has been revised 45 times.

[mruberry]

Cool!

[gchanan]

this is tested, see: #50376.

[facebook-github-bot]

@gchanan merged this pull request in b001c4c.