add NA helpers

[espinielli]

Add functions like in tidyverse/tidyr#950 (comment), namely

dat <- tribble(
  ~x, ~y,
  1 ,  2,
  3 , NA,
  NA, NA
)

## drop rows where any selected column is NA (drop_any)
dat %>% filter(if_all(everything(), ~ !is.na(.x)))

## drop rows where all selected columns are NA (drop_all)
dat %>% filter(if_any(everything(), ~ !is.na(.x)))

## keep rows where any selected column is NA (keep_any)
dat %>% filter(if_any(everything(), is.na))

## keep rows where all selected columns are NA (keep_all)
dat %>% filter(if_all(everything(), is.na))