added custom return object

[puzzleshark]

Restores behavior of pennylane returning custom items to as it was here #1109 (comment)

In this pr, the return object is still an autograd numpy array (which we require for differentiation) with a single entry (being the custom object) and so requires .item() to unpack (see the discussion linked above).

[github-actions]

Hello. You may have forgotten to update the changelog!
Please edit doc/releases/changelog-dev.md with:

• A one-to-two sentence description of the change. You may include a small working example for new features.
• A link back to this PR.
• Your name (or GitHub username) in the contributors section.

[codecov]

Codecov Report

Merging #2808 (84b135f) into master (465642b) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master    #2808   +/-   ##
=======================================
  Coverage   99.63%   99.63%           
=======================================
  Files         252      252           
  Lines       20986    20989    +3     
=======================================
+ Hits        20910    20913    +3     
  Misses         76       76           

Impacted Files	Coverage Δ
pennylane/_qubit_device.py	99.52% <100.00%> (+<0.01%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 465642b...84b135f. Read the comment docs.

[antalszava]

This is looking great overall @puzzleshark! 🎉 Some minor comments.

Also, don't forget to update the changelog! :)

[cvjjm]

When running this test with the latest master merged in I get an additional error:

        with qml.tape.Unwrap(*tapes):
            res, jacs = execute_fn(tapes, **gradient_kwargs)

        for i, r in enumerate(res):

            if isinstance(r, np.ndarray):
                # For backwards compatibility, we flatten ragged tape outputs
                # when there is no sampling
                try:
>                   if isinstance(r[0][0], dict):
E                   TypeError: 'SpecialObject' object is not subscriptable

pennylane/interfaces/autograd.py:118: TypeError

To fix this TypeError probably needs to be added to the try block in autograd.py

[antalszava]

Looking good overall @puzzleshark, thank you for looking into this! 💯

The main change to do from my side would be moving the test cases into another more relevant file.

[antalszava]

Looks good! 🎉 Great one 🙂