This reverts a334425c.

The main reason is that now the workflow is inconsistent when using
spring.

When using spring `RAILS_ENV` is always set, so only one database is
created.

This means that in development `bin/rake db:create` and `bundle exec
rake db:create` have different results.

It also breaks the `bin/setup` script since `bin/rake db:setup
db:test:prepare` will fail.

We should use #find_or_initialize_by and #find_or_create_by because #first_or_initialize and #first_or_create methods are not the public API

```
$ ARCONN=mysql2 be ruby -w -Itest test/cases/scoping/default_scoping_test.rb
Using mysql2
/Users/kamipo/src/github.com/rails/rails/activerecord/lib/active_record/scoping/named.rb:158: warning: method redefined; discarding old female
/Users/kamipo/src/github.com/rails/rails/activerecord/lib/active_record/scoping/named.rb:158: warning: previous definition of female was here
/Users/kamipo/src/github.com/rails/rails/activerecord/lib/active_record/scoping/named.rb:158: warning: method redefined; discarding old male
/Users/kamipo/src/github.com/rails/rails/activerecord/lib/active_record/scoping/named.rb:158: warning: previous definition of male was here
```

I'm unsure how this passed CI in the pull request.

Before:

```ruby
create_table "big_numbers", force: :cascade do |t|
  t.integer "bigint_column", limit: 8
end
```

After:

```ruby
create_table "big_numbers", force: :cascade do |t|
  t.bigint "bigint_column"
end
```

The comments of add_foreign_key method was displaying incorrect constraint name.

References https://github.com/rails/homepage/issues/46.

Follow up to #24079

Currently the results of `column.serial?` is not correct. For
`column.serial?` correctly working, initialize `column.table_name`
immediately.

related to #19324 and #18606

[skip ci]

[ci skip]

Follow up to #22543.

There is a need to pass the keyword arguments to `attribute` method.

Since we are using `datetime` in migration, better to use `datetime` is I think that confusion is less.

Currently use minitest mock, mocha is not used.

Introduce a predicate method that doesn't need to build a scope chain,
but also hides the data structure used for representing the scope chain.

We know a reflection can be considered as `nested?` if the source
reflection or the through reflection are also through reflections (since
the through reflection will also add another join table).  This allows
us to avoid traversing the entire reflection tree just to calculate whether or
not there will be more join tables.

Before:

```ruby
create_table "postgresql_big_serials", force: :cascade do |t|
  t.bigserial "seq", limit: 8, null: false
end
```

After:

```ruby
create_table "postgresql_big_serials", force: :cascade do |t|
  t.bigserial "seq", null: false
end
```

Currently column options handled by the type map in Rails 4.2.

Closes #23934.

This is a forward port of ac832a43

Previously the scope of all associations with extensions were wrapped in
an instance dependent proc. This made it impossible to preload such
associations.

See the end of [this] page.

[ci skip]

[this]: http://edgeapi.rubyonrails.org/classes/ActiveRecord/Persistence/ClassMethods.html#method-i-create

Originally, `{insert|update|delete}_sql` is protected methods.
We can use the `{insert|update|delete}` public methods instead.

Simply it is sufficient to use the method in the super class.

These should allow external code to run blocks of user code to do
"work", at a similar unit size to a web request, without needing to get
intimate with ActionDipatch.

When the `source_type` option is passed to a has_many through, the
resulting `Reflection` will be an instance of `PolymorphicReflection`
and not `ThroughReflection`, meaning that it will ignore the scopes that
it needs to apply from the through reflections. This adds an additional
delegation call to remedy this. I've been finding the reflection code
completely impenetrable lately, it could use some major love.

Fixes #22726

Actually `:name` and `:type` are not column options.