Backport of 'Add IP fallback errors to Gitlab::HTTP HTTP_ERRORS' to 19.3

What does this MR do and why?

Backport of !259565 (merged) to 19.3, as requested in !259565 (comment 3986613490).

• !239733 (merged) (shipped in 19.3) made Gitlab::HTTP try the next resolved IP on Errno::EADDRNOTAVAIL, Errno::EAFNOSUPPORT and Errno::ETIMEDOUT. When the last IP also fails, that error is raised, but it is not in HTTP_ERRORS, so callers that rescue HTTP_ERRORS do not catch it.
• Customer impact: on self-managed instances without outbound internet access, users with a GitLab Duo seat get about 10 seconds added to every page. The usage quota check to CustomersDot fails with Errno::EAFNOSUPPORT, the error is not rescued, so the result is never cached and the check repeats on every page. With this fix the failure is rescued and cached for an hour, as on 19.2.
• This backport contains only the gems/gitlab-http change and its spec. The master MR also changed ee/lib/gitlab/subscription_portal/cdot_response_error.rb, which does not exist on 19.3.

MR acceptance checklist

This checklist encourages us to confirm any changes have been analyzed to reduce risks in quality, performance, reliability, security, and maintainability.

• This MR is backporting a bug fix, documentation update, or spec fix, previously merged in the default branch.
• The MR that fixed the bug on the default branch has been deployed to GitLab.com (not applicable for documentation or spec changes).
• The MR title is descriptive (e.g. "Backport of 'title of default branch MR'"). This is important, since the title will be copied to the patch blog post.
• Required labels have been applied to this merge request
  • severity label and bug subtype labels (if applicable)
  • If this MR fixes a bug that affects customers, the customer label has been applied.
• This MR has been approved by a maintainer (only one approval is required).
• Ensure the E2E Omnibus GitLab EE pipeline (triggered by thee2e:test-on-omnibus-ee job) has succeeded. If it has failed (or shows warnings), investigate the failures. If you determine the failures are unrelated, you may proceed. If you need assistance investigating, request help in the #s_developer_experience Slack channel to confirm the failures are unrelated to the merge request.

Note to the merge request author and maintainer

If you have questions about the patch release process, please:

• Refer to the patch release runbook for engineers and maintainers for guidance.
• Ask questions on the #releases Slack channel (internal only).
• Once the backport has been merged, the commit changes will be automatically deployed to a release environment that can be used for manual validation. See after merging runbook for details.