`fast_strptime()` and other parsers should recycle the `isdst` field

[DavisVaughan]

Right now, the parsers use .mklt() which just sets the isdst field to -1L

> lubridate:::.mklt
function (dtlist, tz) 
{
    na_fill <- rep_len(NA_integer_, length(dtlist$sec))
    dtlist[["wday"]] <- na_fill
    dtlist[["yday"]] <- na_fill
    dtlist[["isdst"]] <- -1L
    .POSIXlt(dtlist, tz = tz)
}

This means that this field isn't recycled to the right length but all the others are:

library(lubridate, warn.conflicts = FALSE)

x <- c("68-05-17", "69-05-17", "55-05-17")

# `isdt` should be vectorized
unclass(fast_strptime(
  x, 
  format = "%y-%m-%d",
  tz = "UTC"
))
#> $sec
#> [1] 0 0 0
#> 
#> $min
#> [1] 0 0 0
#> 
#> $hour
#> [1] 0 0 0
#> 
#> $mday
#> [1] 17 17 17
#> 
#> $mon
#> [1] 4 4 4
#> 
#> $year
#> [1] 168  69 155
#> 
#> $wday
#> [1] NA NA NA
#> 
#> $yday
#> [1] NA NA NA
#> 
#> $isdst
#> [1] -1
#> 
#> attr(,"tzone")
#> [1] "UTC"

# It is vectorized in strptime()
unclass(strptime(
  x, 
  format = "%y-%m-%d",
  tz = "UTC"
))
#> $sec
#> [1] 0 0 0
#> 
#> $min
#> [1] 0 0 0
#> 
#> $hour
#> [1] 0 0 0
#> 
#> $mday
#> [1] 17 17 17
#> 
#> $mon
#> [1] 4 4 4
#> 
#> $year
#> [1] 168  69 155
#> 
#> $wday
#> [1] 4 6 1
#> 
#> $yday
#> [1] 137 136 136
#> 
#> $isdst
#> [1] 0 0 0
#> 
#> attr(,"tzone")
#> [1] "UTC"

This causes issues in vctrs with vec_slice():

library(lubridate, warn.conflicts = FALSE)

x <- c("68-05-17", "69-05-17", "55-05-17")

lt <- fast_strptime(
  x, 
  format = "%y-%m-%d",
  tz = "UTC"
)

# The lack of vectorization breaks vctrs subsetting
vctrs::vec_slice(lt, 1)
#> Error in `vctrs::vec_slice()`:
#> ! Column `isdst` (size 1) must match the data frame (size 3).
#> ℹ In file 'slice.c' at line 191.
#> ℹ This is an internal error in the rlang package, please report it to the
#>   package authors.

I think it is reasonable to assume that this field should have been recycled to the length of the vector.

I'll do a PR