fix(instrumentation-fetch): removes CORS sub-span creation as it was not compliant with ResourceTiming spec

[mariomc]

Which problem is this PR solving?

Fixes #5122

In instrumentation-fetch and instrumentation-xhr, we're including logic to find preflight requests and create (nested) spans for them.
However, that is unnecessary complexity as OPTIONS/preflight calls for a cross-domain request are not added as an additional entry in the PerformanceEntry list and, much like redirects and other preconditions, are accounted for in the main resource entry (despite not being able to differentiate between one and the other).

A simple test can be done by executing the following:

(async () => {
  const CORS_URL = "https://httpbin.org/headers";
  await fetch(CORS_URL).then((a) => a.json());
  const entries = performance.getEntriesByName(CORS_URL);
  console.log("Number of entries:", entries.length);
})();

Short description of the changes

Removed CORS child span creation, filtering and other logic.

Type of change

Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)

How Has This Been Tested?

Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration

• Tested the demo for both instrumentation-fetch and instrumentation-xhr in: Google Chrome 130, Safari 18, Firefox 131. Noticed there were no child spans being created for OPTIONS/preflight requests nor performance entries being added for CORS-enabled requests.
• Tested a simple html app doing a cross-domain fetch request and consulted the performance entries.

Checklist:

• Followed the style guidelines of this project
• Unit tests have been added
• Documentation has been updated

[linux-foundation-easycla]

The committers listed above are authorized under a signed CLA.

• ✅ login: mariomc / name: Mário Gonçalves (598a57b, 19d233a)

[github-actions]

This PR is stale because it has been open 60 days with no activity. Remove stale label or comment or this will be closed in 14 days.

[chancancode]

@mariomc not sure if you are still working on this, but there have been some new developments since.

I worked on #5282, which completely rewrote the tests. They are now making real fetch() requests using a service worker interceptor, making the tests less susceptible to these kind of issues. As part of that, I also adjusted the tests as necessary to reflect the real-world CORS behavior.

So I think the test changes from this PR is already taken care of (please double check, though). The refactor to remove the runtime code (that does nothing, as you pointed out), is still needed, as I didn't want to touch the implementation while doing a big test refactor.

A slightly unfortunate thing is that, while the same refactor/general approach to would work for the XHR tests, and IMO a useful improvement, I ran out of time to do that myself. So the test suites have different paradigms, which may affect your plan to fix both, but it looks like you already figured out how to work with that anyway.

[mariomc]

@mariomc not sure if you are still working on this, but there have been some new developments since.

I worked on #5282, which completely rewrote the tests. They are now making real fetch() requests using a service worker interceptor, making the tests less susceptible to these kind of issues. As part of that, I also adjusted the tests as necessary to reflect the real-world CORS behavior.

So I think the test changes from this PR is already taken care of (please double check, though). The refactor to remove the runtime code (that does nothing, as you pointed out), is still needed, as I didn't want to touch the implementation while doing a big test refactor.

A slightly unfortunate thing is that, while the same refactor/general approach to would work for the XHR tests, and IMO a useful improvement, I ran out of time to do that myself. So the test suites have different paradigms, which may affect your plan to fix both, but it looks like you already figured out how to work with that anyway.

I haven't kept up with the latest developments, but will keep up with them once I have an opening.
If you want to pick this up before that, be my guest! 😄

[github-actions]

This PR is stale because it has been open 60 days with no activity. Remove stale label or comment or this will be closed in 14 days.

[pichlermarc]

Closing this PR, as this was already addressed IIUC - please open a new PR if this is incorrect. Thanks! 🙂