Cross-validation with refit=True failing on particular datasets using Pandas

[elephaint]

trafficstars

What happened + What you expected to happen

CV with refit=True on Pandas input DataFrame fails because there seems to be a mismatch in associated dates when the prediciton DataFrame is matched to the test DataFrame here. I think it's because utilsforecast is creating the wrong split datasets, i.e. with the wrong dates, causing the join to fail.

With Polars, it's not an issue. Hence, it's likely that this is a utilsforecast problem, but for bookkeeping sake I'm reporting it here.

Issue was with ILI dataset, I'll post code later.

Versions / Dependencies

3.0.1

Reproduction script

tba

Issue Severity

Medium: It is a significant difficulty but I can work around it.