Analysis results extra setter behavior

[coruscating]

As reported by Erick, after the dataframes refactor, MitigatedTomographyAnalysis isn't able to set the extra flag so the user is unable to easily distinguish whether a result is mitigated or not:

qiskit-experiments/qiskit_experiments/library/tomography/mit_tomography_analysis.py

Lines 110 to 122 in 8cc20df

	for res in tomo_data.analysis_results(block=False):
	res.extra["mitigated"] = True
	# Combine results so that tomography results are ordered first
	combined_data = [tomo_data, roerror_data]
	# Run unmitigated tomography analysis
	if self.options.unmitigated_fit:
	tomo_analysis.set_options(measurement_basis=PauliMeasurementBasis())
	nomit_data = tomo_analysis.run(tomo_data, replace_results=False).block_for_results()
	for res in nomit_data.analysis_results(block=False):
	res.extra["mitigated"] = False
	combined_data.append(nomit_data)

[ItamarGoldman]

With the current version of main, I couldn't recreate the error. @coruscating, could you tag Erick with so I will get a code snippet that recreate the error?
The code I ran was the following:

from qiskit_aer import AerSimulator
from qiskit.providers.fake_provider import FakePerth

backend = AerSimulator.from_backend(FakePerth())

import qiskit
from qiskit_experiments.library import MitigatedStateTomography

# GHZ State preparation circuit
nq = 2
qc_ghz = qiskit.QuantumCircuit(nq)
qc_ghz.h(0)
qc_ghz.s(0)
for i in range(1, nq):
    qc_ghz.cx(0, i)

# QST Experiment
qstexp1 = MitigatedStateTomography(qc_ghz)
qstdata1 = qstexp1.run(backend, seed_simulation=100).block_for_results()

# Print results
for result in mqstdata1.analysis_results():
    print(result)

state_result = mqstdata1.analysis_results("state")
print(state_result.value)

[yaelbh]

@ItamarGoldman I think @nkanazawa1989 is already working on this issue in #1344.