plot_predictions() returns NULL when no CI is provided

[Bisaloo]

example2 <- scoringutils::range_example_data_long

scoringutils::plot_predictions(example2, x = "value_date",
                               filter_truth = list('value_date <= "2020-06-22"',
                                                   'value_date > "2020-05-01"'),
                               filter_forecasts = list("model == 'SIRCOVID'",
                                                       'creation_date == "2020-06-22"'),
                               allow_truth_without_pred = TRUE,
                               facet_formula = geography ~ value_desc)

example3 <- example2[example2$range == 0, ]
scoringutils::plot_predictions(example3, x = "value_date",
                               filter_truth = list('value_date <= "2020-06-22"',
                                                   'value_date > "2020-05-01"'),
                               filter_forecasts = list("model == 'SIRCOVID'",
                                                       'creation_date == "2020-06-22"'),
                               allow_truth_without_pred = TRUE,
                               facet_formula = geography ~ value_desc)
#> NULL

^{Created on 2021-11-22 by the reprex package (v2.0.1.9000)}

In the second example, I would expect a plot without intervals but still a plot. This is a problem in the hub for forecasts that only provide point predictions.