This was caused by 6d0d83a3. While the
bug it's trying to fix is handled if the association is loaded in an
after_(create|save) callback, it doesn't handle any cases that load the
association before the persistence takes place (validation, or before_*
filters). Instead of caring about the timing of persistence, we can just
ensure that we're not double adding the record instead.

The test from that commit actually broke, but it was not because the bug
has been re-introduced. It was because `Bulb` in our test suite is doing
funky things that look like STI but isn't STI, so equality comparison
didn't happen as the loaded model was of a different class.

Fixes #26661.

Fix typo in 5_0_release_notes.md doc [ci skip]

remove `mongrel` once again

Remove dead constants

`mongrel` was removed in #26408. But have back accidentally in #26414.

update description of queue adapter in test [ci skip]

[ci skip] Remove not necessary whitespace

[ci skip] Use right format text

[ci skip] Use class name instead of path to file

[ci skip] Removes broken performance testing link from the 2.2 release notes

`ActiveJob::TestCase` set `TestAdapter` to queue adapter at `before_setup`.
https://github.com/rails/rails/blob/master/activejob/lib/active_job/test_helper.rb#L12..L21

use `active_support/testing/autorun` in `bin/test` script

Make :as option also set request format (AC::TestCase)

Do not leak the Errors default proc when calling to_hash or as_json

29f0fbd5 has changed that always call
`Minitest.autorun` in `active_support/testing/autorun`.

Therefore, when used directly `Minitest.run`, tests are performed twice.

fix typo in `DateAndTime::Calculations#all_week` doc [ci skip]

`Date.week_start` does not exist. `Date.beginning_of_week` seems to be correct.
Ref: #5339

Fix doc format for `RecordFetchWarning` [ci skip]

Initial Rails Commands Infrastructure

This is not as good a solution as actually hashing both values, but Ruby
doesn't expose that capability other than allocating the array. Unless we were
to do something silly like have a thread local array that is re-used, I don't
see any other way to do this without allocation. This solution may not be
perfect, but it should reasonably avoid collisions to the extent that we need.

[ci skip] Update PG adapter documentation

Per discussion in pull request #26622:

"Let's change it to PG::Error. The more specific classes mentioned are
subclasses, and the fact the raised exception is a PG::UndefinedColumn
doesn't change the fact that it's also a PG::Error." - matthewd

[ci skip] Clarify Postgresql Documentation

This clarifies the object that +ActiveRecord::Base.connection.execute+
will return when using Postgresql.

Wash out your old! These adhoc scripts are replaced by the new
commands.

Prescribed some review fixes for myself!

Still vulnerable to different file structures. We likely want something
more robust when we tackle in app commands.