Fix Error in Accuracy Tests for NER Task

[chakravarthik27]

Always getting zero score due to y_true and y_pred are concatenate Example:

y_pred = pd.Series['O', 'ORG', 'O']
y_pred = pd.Series['lLOC', 'ORG', 'O']

Incorrect Extraction of classes:

classes = set(y_true + y_pred) 
classes 
['OLOC', 'ORGORG', 'OO'] 

Fixes in langtest/utils/util_metrics.py

if isinstance(y_true, list) and isinstance(y_pred, list):
    unique_labels = set(y_true + y_pred)
elif isinstance(y_true, pd.Series) and isinstance(y_pred, pd.Series):
    unique_labels = set(y_true.tolist() + y_pred.tolist())

• min_weighted_f1_score
• min_macro_f1_score