51 replace as tibble

[dajmcdon]

Closes #51 .

• Removed the offending as_tibble.epi_df() method.
• Replaced by creating some alternative special processors.

[dshemetov]

After some Slack back and forth, this looks quite good! The summary for me was that the workflow in epipredict looks like

library(epiprocess)
library(dplyr)
library(recipes)

jhu <- jhu_csse_daily_subset %>%
  filter(time_value > "2021-08-01") %>%
  select(geo_value:death_rate_7d_av) %>%
  rename(case_rate = case_rate_7d_av, death_rate = death_rate_7d_av)

r <- epi_recipe(jhu) %>%
  step_epi_lag(death_rate, lag = c(0, 7, 14)) %>%
  step_epi_ahead(death_rate, ahead = 7) %>%
  step_epi_lag(case_rate, lag = c(0, 7, 14)) %>%
  step_naomit(all_predictors()) %>%
  step_naomit(all_outcomes(), skip = TRUE)

workflow <- epi_workflow() %>%
  add_epi_recipe(r)

In that last step, we add default_epi_recipe_blueprint, which has now been customized to not call check_is_data, which contains a call to as_tibble. This dodges our issues with needing to overwrite as_tibble.