Sync style with Basic Authentication and Digest Authentication headers.

[ci skip]

PostgreSQL: Make detection of invalid plan locale-independent

The text of error messages from a PostgreSQL server depend on the
setting "lc_messages" in postgresql.conf.
The current detection of the particular error condition of an invalid
plan therefore works with English language setting on the server only.
Other locale settings result in a wrong exception type, which lead
to two test failures:

HotCompatibilityTest#test_cleans_up_after_prepared_statement_failure_in_a_transaction [test/cases/hot_compatibility_test.rb:75]:
[ActiveRecord::PreparedStatementCacheExpired] exception expected, not
Class: <ActiveRecord::StatementInvalid>
Message: <"PG::FeatureNotSupported: FEHLER:  gecachter Plan darf den Ergebnistyp nicht ändern\n">

Here the German message wasn't recognized, so that an StatementInvalid
is raised instead of PreparedStatementCacheExpired.

Using the PG_DIAG_SOURCE_FUNCTION is independent of the locale setting.
It is not guarantied to be stable, but it has been unchanged since PostgreSQL-9.2.
On the other hand the text to be matched isn't guarantied to be stable either.

* Refactor generator `route` action tests

Add `assert_routes` helper, which verifies that routes are always added
to the `Rails.application.routes.draw` block, are always indented, and
are always terminated with a newline.  This "fixes" one test that was
not testing something it claimed to, and obsoletes another test.

Also, add a test case for indentation of multi-line routing code.

* Add :namespace option to generator `route` action

Modify respond_to behaviour always setting the response's content type based on the request format

Eager load translations during initialization

Fix connection pools not shared between writer -> replica during tests:

Fix IRB deprecation warning on tab-completion on Ruby <= 2.5:

Modify ActiveRecord::TestFixtures to not rely on AS::TestCase:

Update fragment caching doc [ci skip]

- Similar fix as https://github.com/rails/rails/pull/37100/ which
  solved the issue only for ruby 2.6.

  Fix #37775

- ### Problem

  Connection pools are not properly shared. A replica can't see the
  data in the open transaction on the writing connection.
  ```ruby
  Dog.create!(name: 'bilou')

  AnimalsBase.connected_to(role: :reading) do
    Dog.find_by(name: 'bilou') # No result
  end
  ```

  The reason of this bug is because when test starts, we iterate
  over `AR::Base.connection_handlers` which only contains the `writing`
  handler since models haven't been loaded (app isn't eagerloaded
  test env).

  ### Solution

  Share connection pools as soon as a connection is established
  by creating a subcriber.

- ### Problem

  If one wants to use ActiveRecord::TestFixtures it is mandatory for
  the test suite to inherit from `ActiveSupport::TestCase`.
  TestFixtures makes use of specific method from AS::TestCase
  (`file_fixture_path` and `method_name`).

  ### Solution

  This PR fixes that by not making use of method_name and file_fixture_path.

Fix unscoped grouped where

Fix RSpec capitalization typo ("Rspec" -> "RSpec")

[ci skip]

RSpec is written with a capital R and S elsewhere in the guide, change
it here to match.

[regresion] Allow using mountable engine route helpers in System Tests (v6.0.1)

Add comment to BROWSER_LIKE_ACCEPTS regex

activerecord: Add missing require for compact_blank

Prevent nesting failure screenshots into dirs

Remove an unused connection handler in a test

Update guide missing "HTTP" of "HTTP Token authentication" [skip ci]

- `respond_to any` doesn't allow to specify a content type and
  the content type in the response will be based on the request
  format.

  ```ruby
    def my_action
      respond_to do |format|
        format.html { render(html: 'hello') }
        format.any { render(json: { foo: 'bar'}) }
      end
    end

    get('my_action.csv')
    # Before this patch, content type was `text/csv'
    # Ather this patch, content type is correctly set to whateve we did in the `format.any` block
  ```

  If the client specify the type of data he wants but the server
  doesn't know how to handle it and return plain text (or whatever)
  I don't think it make sense to falsey claim that we are returning
  a `text/csv` a response where in fact we are returning something else.

  Fix #37345

There is an unused connection handler in a test and an extra
connection handler is made, so testing for sqlite3 makes an empty db
file as below.

```
$ bundle exec rake test:sqlite3
(snip)
$ git status
On branch test
Untracked files:
  (use "git add <file>..." to include in what will be committed)

        db/
```

Fix broken check on GitHub actions

Without it, it seems to be an error trying to install an older version.
Ref: https://github.com/rails/rails/commit/ea303d012e6638c99f528c68ee9144a83e836e27/checks?check_suite_id=321223550#step:4:29

* Allow system tests using Rack::Test to run without Chrome

If you require "action_dispatch/system_test_case", the driven_by method
will be executed immediately.

* https://github.com/rails/rails/blob/v6.0.0/actionpack/lib/action_dispatch/system_test_case.rb#L162

Then the preload method is called in SystemTesting::Driver#initialize.

* https://github.com/rails/rails/blob/v6.0.0/actionpack/lib/action_dispatch/system_testing/driver.rb#L13
* https://github.com/rails/rails/blob/v6.0.0/actionpack/lib/action_dispatch/system_testing/browser.rb#L46-L63

Therefore, a "Webdrivers::BrowserNotFound" error occurs in the browser
preloading when you run system tests witout Chrome.

This commit avoid the error by lazy configuring the driver.

ref: #37410

* Don't preload the browser with :rack_test

If users specify `driven_by(:rack_test)`, it uses Chrome by default arguments.
However, `Rack::Test` does not use a browser and does not need to be preloaded.

Furthermore, it occurs `Webdrivers::BrowserNotFound` when run in
a container (or a machine) without Chrome.

Add document for including and excluding methods[ci skip]

and `Array#including` and `Array#excluding`