Fix corrupt transaction state caused by `before_commit` exceptions

Solves personal insanity, since I was down in these files.

(Please don't submit PRs like this.)

Some places were saying filter, while others said filter_options, spare the ambiguity
and use filter throughout.

This inlines a needless local variable and clarifies a route filter consists of defaults
and values to match against.

When a `before_commit` callback raises, the database is rolled back but
AR's record of the current transaction is not, leaving the connection in
a perpetually broken state that affects all future users of the
connection: subsequent requests, jobs, etc. They'll think a transaction
is active when none is, so they won't BEGIN on their own. This manifests
as missing `after_commit` callbacks and broken ROLLBACKs.

This happens because `before_commit` callbacks fire before the current
transaction is popped from the stack, but the exception-handling path
they hit assumes that the current transaction was already popped. So the
database ROLLBACK is issued, but the transaction stack is left intact.

Common cause: deadlocked `#touch`, which is now implemented with
`before_commit` callbacks.

What's next:
* We shouldn't allow active transaction state when checking in or out
  from the connection pool. Verify that conns are clean.
* Closer review of txn manager sad paths. Are we missing other spots
  where we'd end up with incorrect txn state? What's the worst that can
  happen if txn state drifts? How can we guarantee it doesn't and
  contain the fallout if it does?

Thanks for @tomafro for expert diagnosis!

Assume the filter is a string, if it wasn't a hash and isn't nil. Remove needless else
and rely on Ruby's default nil return.

Add spaces within hash braces.

Add options for rake routes task

This reverts commit 9f3730a5, reversing
changes made to 2637fb75.

There are additional issues with this commit that need to be addressed
before this change is ready (see #23377). This is a temporary revert in
order for us to have more time to address the issues with that PR,
without blocking the release of beta2.

Address remaining known issues in Interlock

.git added to https repo urls

Add documentation for #17573

Add two options: `-c` and `-g`.
`-g` option returns the urls name, verb and path fields that match the pattern.
`-c` option returns the urls for specific controller.

Fixes #18902, and Fixes #20420

[Anton Davydov & Vipul A M]

git protocol replaced with https

Fixes some parts of #23148.

[ci skip]

Shorten ActiveRecord::InternalMetadata.table_name to ar_internal_metadata

[ci skip] Properly indent code in markdown

[ci skip]

Add Default Puma Config

While we know no user code is running, we should do as much loading as
we can. That way, all the threads will then be able to resume running
user code together.

Otherwise, only the last arriving thread would get to do its load, and
would then return to userspace, leaving the others still blocked.

We mostly care about `reload_classes_only_on_change=true`, because
that's the default... and there, we definitely need to wait for the lock
when necessary.

Change default ACa adapter for development & test

Redis sans EventMachine

for those who already migrated to Rails 5.0.0 beta

to support Oracle database which only supports 30 byte identifier length

Fix documentation related to `config.assets.cache_store` [ci skip]

- sprockets-rails no longer supports customizing cache store after
  rails/sprockets-rails@ecaeb27 using `config.assets.cache_store`.
- Instead we need to configure it using block syntax.
- Fixes #19835.

Added references option to join tables

Avoid extra `show variables` in migration

Fixes issue #22960
When creating join tables with the command

    rails g migration CreateJoinTableShowroomUser showroom:references user:references

The migration will use references to create the joins and output:

     class CreateJoinTableShowroomUser < ActiveRecord::Migration
       def change
         create_join_table :showrooms, :users do |t|
           t.references :showroom, index: true, foreign_key: true
           t.references :user, index: true, foreign_key: true
         end
       end
     end

This allows for proper refrences with indexes and foreign keys to be easily used when
adding join tables. Without `:refrences` the normal output is generated:

    class CreateJoinTableShowroomUser < ActiveRecord::Migration[5.0]
      def change
        create_join_table :showrooms, :users do |t|
          # t.index [:showroom_id, :user_id]
          # t.index [:user_id, :showroom_id]
        end
      end
    end

`initialize_schema_migrations_table` is called in every migrations.

https://github.com/rails/rails/blob/v5.0.0.beta1/activerecord/lib/active_record/migration.rb#L1080
https://github.com/rails/rails/blob/v5.0.0.beta1/activerecord/lib/active_record/schema.rb#L51

This means that extra `show variables` is called regardless of the
existence of `schema_migrations` table.

This change is to avoid extra `show variables` if `schema_migrations`
table exists.

Make to primary key instead of an unique index for internal tables

None of the other components use them, so we should be consistent.

Remove duplicated composite primary key tests

Extract `ExplainPrettyPrinter` to appropriate files

This reverts commit 96355e87, reversing
changes made to a00c36fe.

See https://github.com/rails/rails/pull/23366#issuecomment-177714429