fix(rome_js_formatter): 🐛 preserve new-lines after directives

[IWANABETHATGUY]

Summary

Fix #2437

Test Plan

The updated snapshot should fix related issue #2437,
you could review the snapshot this pr updated.

[IWANABETHATGUY]

!bench_formatter

[github-actions]

Formatter Benchmark Results

group                                    main                                   pr
-----                                    ----                                   --
formatter/checker.ts                     1.00    379.5±1.33ms     6.8 MB/sec    1.02    388.8±1.79ms     6.7 MB/sec
formatter/compiler.js                    1.00    230.8±0.86ms     4.5 MB/sec    1.00    231.0±1.79ms     4.5 MB/sec
formatter/d3.min.js                      1.00    179.1±0.77ms  1498.4 KB/sec    1.00    179.6±0.71ms  1494.4 KB/sec
formatter/dojo.js                        1.00     12.3±0.01ms     5.6 MB/sec    1.00     12.4±0.01ms     5.5 MB/sec
formatter/ios.d.ts                       1.00    277.8±0.52ms     6.7 MB/sec    1.02    282.8±0.58ms     6.6 MB/sec
formatter/jquery.min.js                  1.00     47.2±0.37ms  1794.3 KB/sec    1.00     47.2±0.34ms  1791.5 KB/sec
formatter/math.js                        1.00    362.4±1.17ms  1829.7 KB/sec    1.00    362.9±1.10ms  1826.9 KB/sec
formatter/parser.ts                      1.00      8.6±0.01ms     5.6 MB/sec    1.01      8.6±0.01ms     5.6 MB/sec
formatter/pixi.min.js                    1.00    202.0±0.88ms     2.2 MB/sec    1.00    202.0±0.91ms     2.2 MB/sec
formatter/react-dom.production.min.js    1.00     60.1±0.61ms  1960.7 KB/sec    1.00     59.8±0.77ms  1969.8 KB/sec
formatter/react.production.min.js        1.00      2.9±0.00ms     2.1 MB/sec    1.00      2.9±0.00ms     2.1 MB/sec
formatter/router.ts                      1.00      6.3±0.00ms     9.7 MB/sec    1.01      6.4±0.00ms     9.6 MB/sec
formatter/tex-chtml-full.js              1.00    449.1±1.00ms     2.0 MB/sec    1.00    449.4±1.00ms     2.0 MB/sec
formatter/three.min.js                   1.00    229.8±1.03ms     2.6 MB/sec    1.00    230.2±0.87ms     2.6 MB/sec
formatter/typescript.js                  1.00  1458.5±10.01ms     6.5 MB/sec    1.02   1481.6±3.47ms     6.4 MB/sec
formatter/vue.global.prod.js             1.00     79.3±0.93ms  1555.3 KB/sec    1.00     79.3±0.86ms  1555.0 KB/sec

[IWANABETHATGUY]

It seems that I became a core contributor,
https://github.com/rome/tools/blob/main/CONTRIBUTING.md#if-you-are-a-core-contributor

[ematipico]

It seems that I became a core contributor,
https://github.com/rome/tools/blob/main/CONTRIBUTING.md#if-you-are-a-core-contributor

I think it was a mistake of mine. I didn't know contributors could run bench tests, which is great!

Although compliance tests can't be, that one I know 🤣

[ematipico]

The end result is what we needed! I would suggest the following change.

We already have a logic that checks the presence of lines between nodes:

https://github.com/rome/tools/blob/main/crates/rome_formatter/src/format_element.rs#L828-L844

My suggestion is to make this function public and use this one instead. The reason why we should factor out this function is because it's becoming a recurrent pattern that is used in multiple cases.

[ematipico]

Thank you for your contribution!