Building on @StupidWolf solution, to have all columns:

import numpy as np
import pandas as pd
from IPython.display import display  # Optional
from sklearn.metrics import (
    classification_report,
    precision_recall_fscore_support,
)

res = []
for class_p in classes:
    prec, recall, fbeta_score, support = precision_recall_fscore_support(
        np.array(y_true) == class_p,
        np.array(y_pred) == class_p,
        pos_label=True,
        average=None,
    )
    res.append(
        [
            class_p,
            prec[1],
            recall[1],
            recall[0],
            fbeta_score[1],
            support[1],
        ]
    )

df_res  = pd.DataFrame(
    res,
    columns=[
        "class",
        "precision",
        "recall",
        "specificity",
        "f1-score",∏
        "support",
    ],
)

display(df_res)