sktime: [BUG] `forecaster.update` fails on multi indexed data frame

Discussed in https://github.com/sktime/sktime/discussions/5833

<div type='discussions-op-text'>

^{Originally posted by ManuB68 January 25, 2024} I am trying to update a forecaster but it doesn’t seem to work on my multi indexed data frame:

python code:

df = pd.DataFrame
df = pd.DataFrame([['A', '2024-01-01', 1, 11], ['A', '2024-01-02', 2, 12], ['A', '2024-01-03', 3, 13], ['A', '2024-01-04', 4, 14],\
                   ['B', '2024-01-01', 101, 1], ['B', '2024-01-02', 102, 2], ['B', '2024-01-03', 103, 3], ['B', '2024-01-04', 104, 4],\
                    ['C', '2024-01-01', 10000, 10], ['C', '2024-01-02', 20000, 20], ['C', '2024-01-03', 30000, 30], ['C', '2024-01-04', 40000, 43]],\
                  columns=['Patients', 'Dates', 'M1', 'M2'])

df['Dates']= pd.to_datetime(df['Dates'])

df = df.reset_index(drop = True)
df = df.set_index(["Patients", "Dates"])

y_train, y_test = temporal_train_test_split(df, test_size=1)

fh = 1

forecaster = VAR() 

forecaster.fit(y_train)
y_pred = forecaster.predict(fh=fh)


Day = y_test.index.get_level_values("Dates")[-1]
y_update = y_test[y_test.index.get_level_values("Dates") == Day]

forecaster.update(y_update)

y_pred_updated = forecaster.predict(fh)

Code might look overcomplicated, due to extraction from a code iterating on dates and my poor knowledge of python The execution fails on line: forecaster.update(y_update)

with this message: ValueError: Need at least 3 dates to infer frequency

</div>