Refs #13658

- We have to restore DATABASE_URL to its previous state irrespective of
  previous value is nil or not

Currently Active Record can be configured via the environment variable `DATABASE_URL` or by manually injecting a hash of values which is what Rails does, reading in `database.yml` and setting Active Record appropriately. Active Record expects to be able to use `DATABASE_URL` without the use of Rails, and we cannot rip out this functionality without deprecating. This presents a problem though when both config is set, and a `DATABASE_URL` is present. Currently the `DATABASE_URL` should "win" and none of the values in `database.yml` are used. This is somewhat unexpected to me if I were to set values such as `pool` in the `production:` group of `database.yml` they are ignored.

There are many ways that active record initiates a connection today:

- Stand Alone (without rails)
  - `rake db:<tasks>`
  - ActiveRecord.establish_connection
 
- With Rails
  - `rake db:<tasks>`
  - `rails <server> | <console>`
  - `rails dbconsole`


We should make all of these behave exactly the same way. The best way to do this is to put all of this logic in one place so it is guaranteed to be used.

Here is my prosed matrix of how this behavior should work:

```
No database.yml
No DATABASE_URL
=> Error
```

```
database.yml present
No DATABASE_URL
=> Use database.yml configuration
```

```
No database.yml
DATABASE_URL present
=> use DATABASE_URL configuration
```

```
database.yml present
DATABASE_URL present
=> Merged into `url` sub key. If both specify `url` sub key, the `database.yml` `url`
   sub key "wins". If other paramaters `adapter` or `database` are specified in YAML,
   they are discarded as the `url` sub key "wins".
```

### Implementation

Current implementation uses `ActiveRecord::Base.configurations` to resolve and merge all connection information before returning. This is achieved through a utility class: `ActiveRecord::ConnectionHandling::MergeAndResolveDefaultUrlConfig`.

To understand the exact behavior of this class, it is best to review the behavior in activerecord/test/cases/connection_adapters/connection_handler_test.rb though it should match the above proposal.

This reverts commit 026d0555.

Conflicts:
	activerecord/lib/active_record/fixtures.rb

Fixes #13383

Closes #13623.

Closes #13625

  - Earlier, change_table was creating database-agnostic object.
  - After this change, it will create correct object based on current
    database adapter.
  - This will ensure that create_table and change_table will get same objects.
  - This makes update_table_definition method public and nodoc.
  - Fixes #13577 and #13503

lib/active_record/store.rb:79: warning: method redefined; discarding old color=
lib/active_record/store.rb:79: warning: previous definition of color= was here
lib/active_record/store.rb:83: warning: method redefined; discarding old color
lib/active_record/store.rb:83: warning: previous definition of color was here

Renaming the test accordingly to its behaviour

Adding 'Fixes' statement to changelog

Improving tests legibility & changelog

Undoing mistakenly removed empty line & further improving changelog

has_many definitions with "name" as singular and as plural e.g.
  has_many :welcome_posts_with_comment
  has_many :welcome_posts_with_comments

Ruby mentions it with:

lib/active_record/associations/builder/collection_association.rb:65:
  warning: method redefined; discarding old welcome_posts_with_comment_ids
lib/active_record/associations/builder/collection_association.rb:65:
  warning: previous definition of welcome_posts_with_comment_ids was here
lib/active_record/associations/builder/collection_association.rb:75:
  warning: method redefined; discarding old welcome_posts_with_comment_ids=
lib/active_record/associations/builder/collection_association.rb:75:
  warning: previous definition of welcome_posts_with_comment_ids= was here

The line was duplicated

Previously, this would give an `ArgumentError`:

   class Issue < ActiveRecord::Base
     enum :status, [:open, :finished]
   end

   Issue.open.build # => ArgumentError: '0' is not a valid status
   Issue.open.create # => ArgumentError: '0' is not a valid status

PR #13542 muted the error, but the issue remains. This commit fixes
the issue by allowing the enum value to be written directly via the
setter:

   Issue.new.status = 0 # This now sets status to :open

Assigning a value directly via the setter like this is not part of the
documented public API, so users should not rely on this behavior.

Closes #13530.

These methods were only used for the `IdentityMap` which was removed.
They are no longer used internally and should be removed without replacement.

As they were not `:nodoc:`'ed it's better to deprecate them before removal.

* Move check from generated helper to test_help.rb, so that all
  applications can benefit
* Rather than just raising when the test schema has pending migrations,
  try to load in the schema and only raise if there are pending
  migrations afterwards
* Opt out of the check by setting
  config.active_record.maintain_test_schema = false
* Deprecate db:test:* tasks. The test helper is now fully responsible
  for maintaining the test schema, so we don't need rake tasks for this.
  This is also a speed improvement since we're no longer reloading the
  test database on every call to "rake test".

Add a mention about the automatic generation of scopes based on the
allowed values of the field on the enum documentation.

Previously, the writer methods would simply check whether the passed
argument was the symbol representing the integer value of an enum field.
Therefore, it was not possible to specify the numeric value itself but
the dynamically defined scopes generate where clauses relying on this
kind of values so a chained call to a method like `find_or_initialize_by`
would trigger an `ArgumentError`.

Reference #13530

to make this grant statement described in the document works

GRANT ALL PRIVILEGES ON inexistent_activerecord_unittest.* to 'rails'@'localhost';

- since `Array()` calls `to_ary` or `to_a` on a subject
- the intent is to 'wrap' subject into an array

Outer joins were being built on the root relation klass rather than the
one specified in the join dependency root

https://github.com/rails/rails/commit/2075f39d726cef361170218fd16421fc52bed5a8 introduced a regression in includes/preloades
by calling `read_attribute` on an association when preloading takes places, instead of using loaded records in `association.target`.

tl;dr

Records are not made properly available via `read_attribute` when preloding in simultaneous,
but value of `@loaded` is already set true, and records concatenated in `association.target` on an association object.
When `@loaded` is true we return an object of `AlreadyLoaded` in preload_for. In `AlreadyLoaded` to return preloaded
records we make wrong use of `read_attribute`, instead of `target` records.

The regression is fixed by making use of the loaded records in `association.target` when the preloading takes place.

Fixes #13437

Currently a developer can pass in a YAML configuration that fully specifies connection information:

```
production:
  database: triage_production
  adapter: password
  pool: 5
```

They can also pass in a string that specifies a connection URL directly to an environment key:

```
production: postgresql://localhost/foo
```

This PR allows the use of both a connection url and specifying connection attributes via YAML through the use of the "url" sub key:

```
production:
  url: postgresql://localhost/foo
  pool: 3
```

This will allow developers to inherit Active Record options such as `pool` from `&defaults` and still use a secure connection url such as `<%= ENV['DATABASE_URL'] %>`. The URL is expanded into a hash and then merged back into the YAML hash. If there are any conflicts, the values from the connection URL are preferred. 

Talked this over with @josevalim

Document the internal interfaces of `ConnectionSpecification::Resolver`

Change method name from `config` to `env` to better match the most common use case.