I'm pretty confused about the addition of this method. The documentation
says that it was intended to allow the removal of values from the
default scope (in contrast to #except). However it behaves exactly the
same as except: https://gist.github.com/jonleighton/7537008 (other than
having a slightly enhanced syntax).

The removal of the default scope is allowed by
94924dc3, which was not a change we
could make until 4.1 due to the need to deprecate things. However after
that change #unscope still gives us nothing that #except doesn't already
give us.

However there *is* a desire to be able to unscope stuff in a way that
persists across merges, which would allow associations to be defined
which unscope stuff from the default scope of the associated model. E.g.

  has_many :comments, -> { unscope where: :trashed }

So that's what this change implements. I've also corrected the
documentation. I removed the guide references to #except as I think
unscope really supercedes #except now.

While we're here, there's also a potential desire to be able to write
this:

  has_many :comments, -> { unscoped }

However, it doesn't make sense and would not be straightforward to
implement. While with #unscope we're specifying exactly what we want to
be removed from the relation, with "unscoped" we're just saying that we
want it to not have some things which were added earlier on by the
default scope. However in the case of an association, we surely don't
want *all* conditions to be removed, otherwise the above would just
become "SELECT * FROM comments" with no foreign key constraint.

To make the above work, we'd have to somehow tag the relation values
which get added when evaluating the default scope in order to
differentiate them from other relation values. Which is way too much
complexity and therefore not worth it when most use cases can be
satisfied with unscope.

Closes #10643, #11061.

We can conditional define the tests depending on the adapter or
connection.

Lets keep the skip for fail tests that need to be fixed.

Allows you to call #unscope on a relation with negative equality operators,
i.e. Arel::Nodes::NotIn and Arel::Nodes::NotEqual that have been generated
through the use of where.not.

Fix in-memory tests

order on the old ones

The previous behavior added a major backward incompatibility since it
impossible to have a upgrade path without major changes on the
application code.

We are taking the most conservative path to be consistent with the idea
of having a smoother upgrade on Rails 4.

We are reverting the behavior for what was in Rails 3.x and,
if needed, we will implement a new API to prepend the order clauses in
Rails 4.1.

Removed tests from deprecated code.

Calling default_scope without a proc will now raise `ArgumentError`.

The previous implementation was necessary in order to support stuff
like:

    class Post < ActiveRecord::Base
      default_scope where(published: true)
      scope :ordered, order("created_at")
    end

If we didn't evaluate the default scope at the last possible moment
before sending the SQL to the database, it would become impossible to
do:

    Post.unscoped.ordered

This is because the default scope would already be bound up in the
"ordered" scope, and therefore wouldn't be removed by the
"Post.unscoped" part.

In 4.0, we have deprecated all "eager" forms of scopes. So now you must
write:

    class Post < ActiveRecord::Base
      default_scope { where(published: true) }
      scope :ordered, -> { order("created_at") }
    end

This prevents the default scope getting bound up inside the "ordered"
scope, which means we can now have a simpler/better/more natural
implementation of default scoping.

A knock on effect is that some things that didn't work properly now do.
For example it was previously impossible to use #except to remove a part
of the default scope, since the default scope was evaluated after the
call to #except.

See #9869 and #9929.

The problem arises from the following example:

    class Project < ActiveRecord::Base
      scope :completed, -> { where completed: true }
    end

    class MajorProject < Project
    end

When calling:

    MajorProject.where(tasks_count: 10).completed

This expands to:

    MajorProject.where(tasks_count: 10).scoping {
      MajorProject.completed
    }

However the lambda for the `completed` scope is defined on Project. This
means that when it is called, `self` is Project rather than
MajorProject. So it expands to:

    MajorProject.where(tasks_count: 10).scoping {
      Project.where(completed: true)
    }

Since the scoping was applied on MajorProject, and not Project, this
fails to apply the tasks_count condition.

The solution is to make scoping apply across STI classes. I am slightly
concerned about the possible side-effects of this, but no tests fail and
it seems ok. I guess we'll see.

The file name should be name_scoping_test.rb and the class should be
`NamedScopingTest` according to ActiveRecord::Scoping::Name

The scoping/default.rb and scoping/named.rb got moved under scoping/ in
commit 2b22564c,
but the tests never did.