Use DOM's post-insertion steps for <script> elements

[domfarolino]

This PR (part 1 of probably 2) uses the newly-introduced DOM Standard "post-insertion steps" (see whatwg/dom#1261), which are run for all nodes in a batch of freshly-inserted nodes, after all DOM insertions take place. The purpose of these steps is to provide an opportunity for script executing side effects to take place during the insertion flow, but after after all DOM mutations are completed atomically. This PR effectively supersedes a portion of #4354.

Before this PR, the HTML standard executed scripts during the <script> HTML element insertion steps. This means that when a batch of script elements were "atomically" inserted, each script would run synchronously after its DOM insertion and before the next DOM insertion took place.

After this PR, to make progress on whatwg/dom#808 and move to a more "atomic" model where script execution only takes place after all pending DOM tree insertions happen, script execution moves to a model that more closely resembles that of Chromium and Firefox. We push script execution back to the post-insertion steps, which run after all DOM insertions are complete. This gives two notable observables differences from the old spec:

1. All text nodes atomically inserted as children to a script will run when their parent script executes. Imagine you have an empty parser-inserted script element. Before this PR, doing script.append(new Text("..."), new Text("..."), ...) would "prepare" and "execute" the script synchronously after the first text node was inserted, because HTML says that whenever a child element is inserted to a connected script element, that's when the script gets prepared. All subsequent text nodes that are appended as children would come after the script has already started, so they would not execute. What I think the standard really means though, is to count on the script's "children changed steps" that DOM offers. After this PR, the execution of script is run after the entire batch of children get inserted, because the execution is tied to the "children changed steps", which run after all nodes are inserted. Therefore, execution would include all text that was atomically added to the script. This is tested here; Safari is the odd one out (I guess because it's not using the "children changed steps" properly?)
2. The post-insertion steps run after a parent's "children changed steps" run. This means any nested <script> elements inserted as children to a parent <script> element will run (as a result of its "post-insertion steps") after the parent script gets a chance at running (as a result of its "children changed steps", which run before any post-insertion steps). See this concrete example I added in the spec. This has a WPT too, but it assumes a blend of the old a new models (which forced nested script execution before parent script execution). I'm updating it to match this spec, in DOM: Correct script-in-script atomic move WPT expectations web-platform-tests/wpt#45094.

• At least two implementers are interested (and none opposed):
  • Chrome
  • Firefox
• Tests are written and can be reviewed and commented upon at:
  • https://wpt.fyi/results/dom/nodes/insertion-removing-steps?label=experimental&label=master&aligned
  • Once approved (i.e., we agree on this model), I will make all tests non-tentative
• Implementation bugs are filed:
  • Chromium: TODO
  • Gecko: TODO
  • WebKit: TODO
• MDN issue is filed: Seems like an edge case, so I think we don't need to do anything here.
• The top of this comment includes a clear commit message to use.

(See WHATWG Working Mode: Changes for more details.)

---

/infrastructure.html ( diff )
/scripting.html ( diff )

[domfarolino]

I'm just following the lead of the insertion steps, where we override them once for this specification, and fan them out into element-specific ones. I notice we don't do that for "children changed steps" though (each element just uses that dfn directly, with no HTML middle-man). Not sure what's more desirable?

[domfarolino]

/cc @noamr @mfreed7

[annevk]

It would be good if @smaug---- reviewed this as well.