Compatibility with dplyr 1.1.0

[DavisVaughan]

dplyr 1.1.0 is on its way to CRAN right now.

The checks run by CRAN noticed that eHDPrep is broken by this dplyr release, and it somehow slipped through our own reverse dependency checks.

• na_if() now casts y to the type of x and requires that y be a compatible type. A character "NA" is not a compatible type with a list-column. It seems like you should be using list("NA") here instead, so I've changed to that.
• We have also deprecated usage of ... in across() and instead encourage using anonymous functions. Those were only throwing warnings but I went ahead and updated those too

[tommtoner]

Thanks, Davis, for finding and addressing this issue. I believe that the type change in na_if() produces the error below when running the examples for identify_inconsistency() if using dplyr 1.0.10 (see this check). I have increased the minimum dplyr version required by eHDPrep to 1.1.0 to fix this.

  Error in `dplyr::mutate()`:
  ! Problem while computing `lgl_values = dplyr::na_if(.data$lgl_values,
    list("NA"))`.
  Caused by error in `x == y`:
  ! comparison of these types is not implemented
  Backtrace:
       ▆
    1. ├─eHDPrep::identify_inconsistency(beans, bean_rules)
    2. │ └─... %>% ...
    3. ├─dplyr::mutate(...)
    4. ├─dplyr:::mutate.data.frame(., lgl_values = dplyr::na_if(.data$lgl_values, list("NA")))
    5. │ └─dplyr:::mutate_cols(.data, dplyr_quosures(...), caller_env = caller_env())
    6. │   ├─base::withCallingHandlers(...)
    7. │   └─mask$eval_all_mutate(quo)
    8. ├─dplyr::na_if(.data$lgl_values, list("NA"))
    9. └─base::.handleSimpleError(...)
   10.   └─dplyr (local) h(simpleError(msg, call))
   11.     └─rlang::abort(...)
  Execution halted```