Improve documentation

Closes #15024.

These operations happen before instances are created.
The current behavior is misleading and can result in broken behavior.

When scanning the Enum API page I found it easy to miss that scopes are generated.  Adding an example clarifies use and has more... *pop*...

We are reverting these commits, because there are other caveats related
to dirty attributes not being restored when a transaction is rollback.
For instance, nested transactions cannot proper restore the dirty
attributes state after a rollback.

At the moment, the dirty attributes are scoped by the transaction.
When we call `.save` on a record, the dirty attributes will be reset even
if the transaction gets rollback.

[related #13166]
[related #15018]
[related #15016]
[closes #15019]

This reverts commits
* bab48f0a
  * b0fa7cf3.
* 73fb39b6
  * 37c23892.
  * 8d8d4f15

Revert "Merge pull request #13166 from bogdan/transaction-magic"

I wanted the middle model to be anonymous, but we have to give it a name
for marshal support. :'(

HABTM should fall back to using the normal CollectionAssociation's size calculation if the collection is not cached or loaded.

This addresses both #14913 and #14914 for master.

[related #13166]

Rewrite to avoid 'we'/'you', add missing period, and keep lines at 80 chars. Cheers :)

Improve readability with help from @senny

[ci skip]

Now that delete_all with destroy or delete_all dependency behave
the same we no longer need this conditional. This means we can remove
the new delete_all_with_dependency method I added and go straight
to delete_records from delete_all, passing :all and the dependent
directly.

Added #join_id_for(owner) to reflection to avoid accessing the source_macro

This makes table_name_suffix work the same as table_name_prefix when
using namespaced models. Pretty much the same as 67d1cec4.

This is a regression 4.0 -> 4.1 fix.
In 4.1.0 Relation#join is delegated to Arel#SelectManager.
In 4.0 series it is delegated to Array#join

This patch puts back the behaviour of 4.0

The case where the value is present is more common

column_for will raise in case column is not found for the given table,
so there is no need to handle that here.

This is a follow up to #14139.

/cc @carlosantoniodasilva

bind parameters we not being propogated to simple subquery calculation
calls. This fixes it

associations were not being saved.

Fixes #13854.

[Eric Chahin, Aaron Nelson, & Kevin Casey]

For mysql2/mysql adapters, `sql_mode` variable name set in `database.yml`
as string, was ignored and `sql_mode` was set to use strict mode.

Fixes #14895

Nullify (or nil dependency) was doing the same thing delete_all
was doing in issue #14546, creating a large IN statement if
the association was loaded. Loaded and not loaded associations
should behave the same. The IN statement is also not great because
it's inefficient.

After reviewing this code I realized the conditional that was
there previously was basically saying if the dependency is not
delete all. This is a better, cleaner, and clearer way to write
this method.