Add input_nodes attribute

[d1ssk]

Before submitting, please check the following:

• Make sure you have tests for the new code and that test passes (run tox)
• format added code by black -l 120 <filename>
• If applicable, add a line to the [unreleased] part of CHANGELOG.md, following keep-a-changelog.

Then, please fill in below:

Context (if applicable): In the current implementation, the Pattern class does not explicitly possess the attribute of input nodes. Instead, it implicitly assumes that nodes with indices ranging from 0 to (width-1) are the inputs. It is considered preferable to have input_nodes as an attribute to accommodate more general situations.

Description of the change:

• Add input_nodes attribute to the Pattern class and the LocalPattern class
• Add input attributes to the CommandNode class
• Add implementations to set the input_nodes during various pattern generation processes (e.g., transpiling from circuits, generating from graphs).
• Add a leave_inputs option to perform_pauli_measurements() (default: False). When set to True, input nodes will not be removed even if their measurement basis is Pauli. When set to False, indices of removed input nodes will be deleted from the input_nodes attribute.

Related issue:

also see that checks (github actions) pass. If lint check keeps failing, try installing black==22.8.0 as behavior seems to vary across versions.

[shinich1]

Is the intention to merge this before the resource state visualisation?

[d1ssk]

Yes, my intention is to merge this change first, then make a minor update to the resource state visualisation and after that, merge the visualization.

[masa10-f]

@d1ssk The code looks good! Could you add a test for the 'leave_input' option of 'perform_pauli_measurements' into 'test_pattern.py' to automatically verify the new option functions correctly?

[shinich1]

@d1ssk please squash and merge, with commit message add input_nodes attribute to pattern class (#88)