[BUG] null.qubit no longer plays nicely with custom operations

[obliviateandsurrender]

Expected behavior

import pennylane as qml
from pennylane.resource.error import ErrorOperation

class IssueErrorOp(ErrorOperation):
    def __init__(self, phase, wires):
        self.phase = phase
        super().__init__(phase, wires=wires)
    def error(self):
        return 1.0

@qml.qnode(qml.device('null.qubit', wires=2))
def circuit():
    IssueErrorOp(1.0, [0])
    return qml.state()

>>> circuit()
[0.0]

null.qubit should allow for adding custom operators to the workflow.

Actual behavior

import pennylane as qml
from pennylane.resource.error import ErrorOperation

class IssueErrorOp(ErrorOperation):
    def __init__(self, phase, wires):
        self.phase = phase
        super().__init__(phase, wires=wires)
    def error(self):
        return 1.0

@qml.qnode(qml.device('null.qubit', wires=2))
def circuit():
    IssueErrorOp(1.0, [0])
    return qml.state()

>>> circuit()
DeviceError: Operator ReplyErrorOp(1.0, wires=[0]) not supported on null.qubit and does not provide a decomposition.

Additional information

This issue exists on the master and possibly comes from using DefaultQubit's preprocess for null.qubit as well.

Source code

No response

Tracebacks

No response

System information

Name: PennyLane
Version: 0.36.0.dev0
Summary: PennyLane is a cross-platform Python library for quantum computing, quantum machine learning, and quantum chemistry. Train a quantum computer the same way as a neural network.
Home-page: https://github.com/PennyLaneAI/pennylane
Author: 
Author-email: 
License: Apache License 2.0
Location: /Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/site-packages
Editable project location: /Users/utkarsh/Xanadu/Codes/SourceCodes/pennylane
Requires: appdirs, autograd, autoray, cachetools, networkx, numpy, pennylane-lightning, requests, rustworkx, scipy, semantic-version, toml, typing-extensions
Required-by: amazon-braket-pennylane-plugin, PennyLane-Catalyst, PennyLane-Cirq, PennyLane-qiskit, PennyLane_Lightning, quantum-datasets

Platform info:           macOS-13.6.4-arm64-arm-64bit
Python version:          3.10.4
Numpy version:           1.26.4
Scipy version:           1.12.0
Installed devices:
- default.clifford (PennyLane-0.36.0.dev0)
- default.gaussian (PennyLane-0.36.0.dev0)
- default.mixed (PennyLane-0.36.0.dev0)
- default.qubit (PennyLane-0.36.0.dev0)
- default.qubit.autograd (PennyLane-0.36.0.dev0)
- default.qubit.jax (PennyLane-0.36.0.dev0)
- default.qubit.legacy (PennyLane-0.36.0.dev0)
- default.qubit.tf (PennyLane-0.36.0.dev0)
- default.qubit.torch (PennyLane-0.36.0.dev0)
- default.qutrit (PennyLane-0.36.0.dev0)

Existing GitHub issues

• I have searched existing GitHub issues to make sure the issue does not already exist.

[albi3ro]

I have been wanting to extend null.qubit to be able to able to mimic any other device for their preprocessing. We should potentially allow the submission other sorts of alternate preprocessing too.

[trbromley]

Thanks @obliviateandsurrender! CC @DSGuala.

This looks like a nontrivial problem for use of specs and the Tracker when working with custom error ops.