feature: support LinearCombination as observable

[ashlhans]

Issue #, if available:
#245

Description of changes:

• Add support for the new LinearCombination as an observable that will be the default in the new version of PennyLane
• Add support for the qml.opsSProd and qml.ops.Sum in the observable

Testing done:
tox and tox -e integ-tests

Merge Checklist

Put an x in the boxes that apply. You can also fill these out after creating the PR. If you're unsure about any of them, don't hesitate to ask. We're here to help! This is simply a reminder of what we are going to look for before merging your pull request.

General

• I have read the CONTRIBUTING doc
• I used the commit message format described in CONTRIBUTING
• I have updated any necessary documentation, including READMEs and API docs (if appropriate)

Tests

• I have added tests that prove my fix is effective or that my feature works (if appropriate)
• I have checked that my tests are not configured for a specific region or account (if appropriate)

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 100.00%. Comparing base (184745c) to head (6bfb893).

Additional details and impacted files

@@            Coverage Diff            @@
##              main      #246   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files            7         7           
  Lines         1207      1215    +8     
  Branches       290       295    +5     
=========================================
+ Hits          1207      1215    +8     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[speller26]

Rather than doing conditional imports, I'd say it's fine to wait until 0.36 is released and import normally

[astralcai]

When I proposed the solution I did not expect the issue of having to do conditional imports. In that case I would say:

isinstance(..., (qml.ops.Hamiltonian, qml.Hamiltonian))

would suffice. Depending on whether new opmath is enabled, qml.Hamiltonian will point to the correct thing. The only time this would cause an error is if the user is explicitly creating a LinearCombination with new opmath disabled, which seems like an edge case that wouldn't realistically happen (the only practical reason for disabling new opmath is typically to work with the legacy Hamiltonian).

[speller26]

Does this work?

Suggested change

	@_translate_observable.register(qml.ops.Hamiltonian)
	@_translate_observable.register(qml.Hamiltonian)
	def _(H: Union[qml.ops.Hamiltonian, qml.Hamiltonian]):
	@_translate_observable.register
	def _(H: Union[qml.ops.Hamiltonian, qml.Hamiltonian]):

Otherwise,

@_translate_observable.register(qml.ops.Hamiltonian)
@_translate_observable.register(qml.Hamiltonian)
def _(H):

[astralcai]

Only since Python 3.11