Tables in tests are not always empty so `klass.first` does not always
find last created record.

Fixes #36479.

https://buildkite.com/rails/rails/builds/61744#f12cc6cf-7458-4131-917a-9735615f6259/999-1010

Move while_preventing_writes from conn to handler

[ci skip] Update docs as `remove_column` can be reversed

Update multi-db docs

Fixed a couple of typos, word 'deliberately' and database_resolver_context class name. [ci skip]

Fix sentence and a typo

Document multiple databases in Rails

Move SchemaMigration to migration_context

Add support for multiple databases to `rails db:abort_if_pending_migrations`

When a route was defined within an optional scope, if that route didn't
take parameters the scope was lost when using path helpers. This patch
ensures scope is kept both when the route takes parameters or when it
doesn't.

Fixes #33219

[ci skip] Use default path in button_to documentation

* Make ActiveRecord `ConnectionPool.connections` thread-safe.

ConnectionPool documentation is clear on the need to synchronize
access to @connections but also states that public methods do not
require synchronization. Existing code exposed @connections
directly via attr_reader. The fix uses synchronize() to lock
@connections then returns a copy to the caller using Array.dup().

Includes comments on the connections method that thread-safe access
to the connections array does not imply thread-safety of accessing
methods on the actual connections.

Adds a test-case that modifies the pool using a supported method
in one thread  while a second thread accesses pool.connections.
The test fails without this patch.

Fixes #36465.

* Update activerecord/test/cases/connection_pool_test.rb

[jeffdoering + Rafael Mendonça França]

images/getting_started: Update screenshot for missing action

Add missing file to require digest/uuid on active_support core ext

Clear Resolvers' cache after RenderTestCases tests

Allow column name with function (e.g. `length(title)`) as safe SQL string

`split(/\s*,\s*/)` to order args and then `permit.match?` one by one is
much slower than `permit.match?` once.

Add forgotten nodoc to dump_schema method.

Allow quoted identifier string as safe SQL string

Fix preloading on AR::Relation where records are duplicated by a join

Include warning in docs about polymorphism in underlying tables

fix typo in CHANGELOGs.

Move schema cache to pool

Fixed db:prepare task for multiple databases.

Treat ActiveRecord::Base and ApplicationRecord as "primary"

Extract default log file open operation to method

Fix sqlite3 collation parsing when using decimal columns.

Don't call listen with empty directory list

Fix EventedFileUpdateChecker through a symlink 

Do not clear deprecated initializer dependencies if using classic autoloader

`Rails::Application::Finisher` defines a `:let_zeitwerk_take_over` initializer. This initializer is always run but it's statements are wrapped in a guard: `config.autoloader == :zeitwerk`.

`Finisher` also defines a `initializer :warn_if_autoloaded` initializer with a `before: :let_zeitwerk_take_over` option which also always runs. This initializer unloads any constants autoloaded during initialization and displays a deprecation warning. This initializer does not account for `config.autoloader` being set to `:classic`.

The problem is that this initializer changes the behavior of the classic autoloader. Constant autoloading from initializers is deprecated but the deprecation should not break existing applications which currently depend on autoloaded constants in initializers.

This commit prevents the dependencies from being unloaded if the autoloader is not Zeitwerk. It also updates the deprecation warning, if the classic autoloader is enabled, to indicate that the constants would have been unloaded if Zeitwerk had been used.

Only clear cache when view paths are specified

Fix invalid schema dump when primary key column has a comment

Avoid making extra 5 arrays in each `save`

Fix table comment also being applied to the primary key column

Change `ActionDispatch::Response#content_type` returning Content-Type header as it is

Fail parallel tests if workers exit early

The other template files do not add `frozen_string_literal`, so should
behave the same.

Ref: #30342, #30348.