[IDAG] Upstream Multi-Device Selection + Tests

[fknorr]

This upstreams the new device selection logic developed by @psalz for the old multi-GPU Celerity branch alongside modifications I made for the new IDAG backend logic.

While the old single-device selection logic remains in place until the IDAG execution PR, its tests are already replaced here to streamline the merge process.

[github-actions]

Check-perf-impact results: (7b849de16ff11660b98988ab0b032db7)

❓ No new benchmark data submitted. ❓
Please re-run the microbenchmarks and include the results if your commit could potentially affect performance.

[github-actions]

clang-tidy made some suggestions

[github-actions]

warning: the parameter 'i' is copied for each invocation but only used as a const reference; consider making it a const reference [performance-unnecessary-value-param]

Suggested change

	mock_device(impl i) : m_pimpl(std::make_shared<impl>(i)) {}
	mock_device(const impl& i) : m_pimpl(std::make_shared<impl>(i)) {}

[coveralls]

Pull Request Test Coverage Report for Build 9582564578

Details

• 107 of 109 (98.17%) changed or added relevant lines in 2 files are covered.
• 66 unchanged lines in 2 files lost coverage.
• Overall coverage decreased (-1.2%) to 93.536%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
include/device_selection.h	99	100	99.0%
include/utils.h	8	9	88.89%

Files with Coverage Reduction	New Missed Lines	%
src/task.cc	1	87.58%
include/device_queue.h	65	39.23%

Totals
Change from base Build 9549035557:	-1.2%
Covered Lines:	7095
Relevant Lines:	7376

---

💛 - Coveralls

[psalz]

Remarkable, this might be your best work yet! 😜 Jokes aside, the new backend selection logic looks good!

Now that it ended up being a separate commit after all, you could throw me a bone with a Co-authored-by :-)