reduce overhead of to_sparse_data_frame()

[simonpcouch]

Noticed this when reviewing #1167 but it's unrelated to that PR. If we do the "cheap" side of the && first, we can cut down on the overhead of "not converting to sparse when we don't want to." :)

Before this PR:

d <- data.frame(sparse = "no")
lr <- linear_reg()
bench::mark(
  to_sparse_data_frame(d, lr)
)
#> # A tibble: 1 × 13
#>   expression              min median `itr/sec` mem_alloc `gc/sec` n_itr  n_gc total_time result
#>   <bch:expr>           <bch:> <bch:>     <dbl> <bch:byt>    <dbl> <int> <dbl>   <bch:tm> <list>
#> 1 to_sparse_data_fram… 26.9µs 28.5µs    33538.        0B     16.8  9995     5      298ms <df>  
# ℹ 3 more variables: memory <list>, time <list>, gc <list>

Now:

#> # A tibble: 1 × 13
#>   expression              min median `itr/sec` mem_alloc `gc/sec` n_itr  n_gc total_time result
#>   <bch:expr>           <bch:> <bch:>     <dbl> <bch:byt>    <dbl> <int> <dbl>   <bch:tm> <list>
#> 1 to_sparse_data_fram… 16.1µs 17.5µs    54942.        0B     16.5  9997     3      182ms <df> 

We're talking about 1.5% of the total time in fit(lr, mpg ~ ., mtcars) vs 2.5%, but seems like a simple enough change to be worth it!

[EmilHvitfeldt]

nice find!

[github-actions]

This pull request has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.