Fix collision check when using a generator and using the force option

Add ActiveRecord::Validations::NumericalityValidator

Add Active Record numericality validator with support for casting
floats using a database columns' precision value.

Only assign @new_record_before_save once in autosave_association

Use full option names in psql / pg_dump for clarity

If we defined a callback before an association that updates the object, then this may end up being
manipulated to being `false` when it should be `true`. We guard this be only defining it once.

The implication of it being false, in this case, is that the children aren't updated with the parent_id
and so they fail to associate to one another.

See https://github.com/rails/rails/issues/38120 for more details

Define mattr_* methods at caller's location

Avoid unnecessary SQL query when calling ActiveRecord::Relation#cache_key

Add `:round_mode` parameter support to number helpers

Fix the broken anchor of link [ci skip]

Support handling a `round_mode` in `ActiveSupport::NumberHelper::RoundingHelper`

Add default value to locale file

Update inline documentation with new parameter option

Update CHANGELOG

Add round_mode examples to all the tests

Add I18n test

Simplify logic

Further simpification

Include primary key in insert_all conflict target if specified

Before:

    [1] pry(main)> $ ActiveJob::Base.default_priority

    From: lib/active_support/core_ext/module/attribute_accessors.rb @ line 64:
    Owner: ActiveJob::QueuePriority::ClassMethods
    Visibility: public
    Number of lines: 3

    def #{sym}
      @@#{sym}
    end

After:

    [1] pry(main)> $ ActiveJob::Base.default_priority

    From: /home/jhawthorn/src/rails/activejob/lib/active_job/queue_priority.rb @ line 9:
    Owner: ActiveJob::QueuePriority::ClassMethods
    Visibility: public
    Number of lines: 1

    mattr_accessor :default_priority

Previously this was incorrectly defining this attribute on the singleton
class, which would end up actually using the class variable from Module.

https://buildkite.com/rails/rails/builds/66168#b348f446-0406-425e-ba53-c5d15034cf98/1002-1070

Fix adding non-null column to existing SQLite3 table

Forces adding the column by way of copying the table due to lack of
support in SQLite3 adapter.

Fixes #38129

Avoid assigning duplicate values to join_values

The ArgumentError occurs even though structures is compatible.
Because some query methods keep duplicate values.

For example, the behavior of `joins` method is as following:

```ruby
relation = Post.joins(:author).joins(:author)
relation.joins_values
#=> [:author, :author]
relation.or(Post.joins(:author))
#=> ArgumentError: Relation passed to #or must be structurally compatible. Incompatible values: [:joins]
```

This commit changes to not keep duplicate values.

Fixes #38052

Update "The Hard Way" setup for Arch Linux

MariaDB cannot be started on Arch Linux before running a command to create the database files

See https://wiki.archlinux.org/index.php/MariaDB#Installation

It also appears that memcached is not started by default on Arch Linux,
so it has been added to the systemctl startup step.

This PR adds that step when setting up on Arch Linux

Pass env_name as a string in test databases

In 154abcab we switched from using `Rails.env` to fetch the `env_name` to
`ActiveRecord::ConnectionHandling::DEFAULT_ENV.call.to_sym` which
changed the type from a `String` to a `Symbol`.

This commit brings things back to the original state, so we can find the
configurations correctly!

It also modifies the configuration in the configurations array, so that
future connections can find the database with the updated keyword value.

When all IPs are trusted, use the furthest away

Fix regression where AJ#perform_now doesn't return the job's value:

- I made a change in 0d3aec49 to output a log if a job was aborted
  in a before callbacks. I didn't take in consideration that a job
  could return a falsy value and thus it would wrongly log
  that the job was aborted.

  This fixes the problem by checking if the callback chain was halted
  rather than the return value of the job.

- Fix #38040

Scenario: we have a REMOTE_ADDR of `127.0.0.1`, and X-Forwarded-For is
`A, B, C`.

Without any relevant trust, the `remote_ip` is `C`.

If `C` is trusted, then the `remote_ip` is `B`.

If `B` and `C` are trusted, then the `remote_ip` is `A`.

If all of `A`, `B`, and `C` are trusted, then the `remote_ip` should
still be `A`: if our trust was sufficient to get that far out before,
trusting something else should not have us fall back to `127.0.0.1`.

It is this last situation that we're correcting here:

We trust `A` to give us accurate X-Forwarded-For information, yet it has
chosen to leave it unset. Therefore, `A` is telling us that it is itself
the client.

Fix regression, enforce fresh ETag header after collection contents change