Currently `SchemaDumper` is only customizable for column options. But
3rd party connection adapters (oracle-enhanced etc) need to customizable
for table or index dumping also. To make it possible, I introduced
adapter specific `SchemaDumper` classes for that.

ActiveRecord associations automatically guess the inverse associations.
But this feature does not work correctly on assoctions for STI.
For example, before this commit

```
class Post < ActiveRecord::Base
  belongs_to :author
end

class SpecialPost < Post; end

class Author < ActiveRecord::Base
  has_many :posts
  has_many :special_posts
end
```

`author.posts.first.author` works correctly, but
`author.special_posts.first.author` does not work correctly.

`SET time zone 'value'` is an alias for `SET timezone TO 'value'`.

https://www.postgresql.org/docs/current/static/sql-set.html

So if `variables["timezone"]` is specified, it is enough to
`SET timezone` once.

This is only used for the internal `column_spec` and
`column_spec_for_primary_key`.

currently integer types extracts the `limit` from `sql_type`. But the
lookup key of type map is the `oid` in postgresql adapter. So in most
case `sql_type` is passed to `extract_limit` as `""` and `limit` is
extracted as `nil`.

https://github.com/rails/rails/blob/v5.1.0.beta1/activerecord/lib/active_record/connection_adapters/postgresql_adapter.rb#L445

In mysql2 adapter, `limit` is registered correctly without extracting
from `sql_type`.

https://github.com/rails/rails/blob/v5.1.0.beta1/activerecord/lib/active_record/connection_adapters/abstract_mysql_adapter.rb#L678-L682

Postgresql adapter should also be registered correctly.

``` ruby
  conn = ActiveRecord::Base.connection
  conn.select_all("SELECT 1::smallint, 2::integer, 3::bigint").column_types.map do |name, type|
    [name, type.limit]
  end
```

Before:

``` ruby
  # => [["int2", nil], ["int4", nil], ["int8", nil]]
```

After:

``` ruby
  # => [["int2", 2], ["int4", 4], ["int8", 8]]
```

Since 213796fb, it was lost the ability that SQL with binds for `insert`,
`update`, and `delete` (like `select_all`). This restores the ability
because `insert`, `update`, and `delete` are public API, so it should
not be removed without deprecation.

Because `to_sql` is public API. I introduced `to_sql_and_binds` internal
API to return SQL and binds.

[ci skip]

And enable `context_dependent` of Style/BracesAroundHashParameters cop.

`write_attribute_without_type_cast` is defined as a private method in
`AttributeMethods::Write`, but `AttributeMethods::Dirty` overrode it as
a public method. It should be kept the original visibility.

Follow up of #29834.

Fixes #30266.

When using a has_many through relation and then summing an attribute
the distinct was not being used. This will ensure that when summing
an attribute, the number is only used once when distinct has been used.

The primary key on the owner record is abstracted as `join_foreign_key`.

`source_type_info` is only used for `constraints` in
`PolymorphicReflection`.

We already have a test case for `serialize` with a custom coder in
`PostgresqlHstoreTest`.

https://github.com/rails/rails/blob/v5.1.3/activerecord/test/cases/adapters/postgresql/hstore_test.rb#L316-L335

Since `Relation` includes `Enumerable`, it is enough to use `super`
simply.

It is incorrect to treat `options[:primary_key]` as
`association_primary_key` if `has_many` associations because the
`:primary_key` means the column on the owner record, not on the
association record. It will break `ids_reader` and `ids_writer`.

```ruby
  people(:david).essay_ids
  # => ActiveRecord::StatementInvalid: Mysql2::Error: Unknown column 'essays.first_name' in 'field list': SELECT `essays`.first_name FROM `essays` WHERE `essays`.`writer_id` = 'David'
```

Fixes #14439.

The previous paragraph mentions that you can hand off the same processing
queue to multiple workers. This completes the following example below it.

It can use `AbstractReflection#table_name` simply.

This comment was added at 97849deb, but `AssociationProxy` and
`test_triple_equality` was removed at 1644663b. Currently the `===` is
used for `test_decorated_polymorphic_where` that added at #11945.
So I updated "association proxies" to "decorated models".

And also, currently `Core::ClassMethods` appears in the doc.

http://api.rubyonrails.org/classes/ActiveRecord/Core/ClassMethods.html

But it looks like that the methods in the module is not public API.
So I also added `# :nodoc:` to the module.

Fixes #29045.

refs: https://github.com/rails/rails/pull/30161

```
$ echo "+@size+" | rdoc --pipe
<p>+@size+</p>

$ echo "<tt>@size</tt>" | rdoc --pipe
<p><code>@size</code></p>
```

[ci skip]

Otherwise `ConnectionPool#reap` may run before `@connections` has
initialized.

https://travis-ci.org/rails/rails/jobs/263037427#L888-L890

This module has behavior that is not present in `ActiveModel::Dirty`,
which is intended to be public API.

Currently if relation object are passed to where condition for has one or has many association wrong set of primary key and foreign key are selected.


Changed code to use 'join' primary key and foreign key over 'association' primary key and foreign key.

`ActiveRecord::ConnectionAdapters::QueryCache::ConnectionPoolConfiguration`
depends on `Concurrent::Map`.