The perf gain is relatively minor but consistent:

```
Calculating -------------------------------------
             0.zero?   137.091k i/100ms
             1.zero?   137.350k i/100ms
              0 == 0   142.207k i/100ms
              1 == 0   144.724k i/100ms
-------------------------------------------------
             0.zero?      8.893M (± 6.5%) i/s -     44.280M
             1.zero?      8.751M (± 6.4%) i/s -     43.677M
              0 == 0     10.033M (± 7.0%) i/s -     49.915M
              1 == 0      9.814M (± 8.0%) i/s -     48.772M
```

And try! is quite a big hotspot for us so every little gain is appreciable.

* Rename `ActiveSupport::Try` => `ActiveSupport::Tryable`
* Include the modules inline
* `private` indentation

`Delegator` inherits from `BasicObject`, which means that it will not
have `Object#try` defined. It will then delegate the call to the
underlying object, which will not (necessarily) respond to the method
defined in the enclosing `Delegator`.

This patches `Delegator` with the `#try` method to work around the
surprising behaviour.

Fixes #5790

Also add doc examples for `Object.nil`.

[ci skip]

@rafaelfranca suggested in f7c7bcd9 that code examples should display
the result after  `# =>` and not after `#=>`.

This commit replaces *all* the occurrences of `#=>` in the code documentation
(mostly added by me 😭) with the suggested `# =>`.

@carlosantoniodasilva pointed out that when `@person` is nil then this would blow up when you ended up calling `#first`on `nil`.

> "there’s no way to break a try chain when you enter it :D"

[ci skip]

- better `if` example
- Added chaining example to the try method description
- Documented the `respond_to?` check to the try method description
- Clearer wording to explain that argument error is raised on argument mismatch to responding method, rather than to non-responding method (which is handled without exception by `try`)
- `.any?` is more precise than `! .blank?`
 - Don't need to use `try` on `children` as (for regular associations) they will always be a collection or array that responds to `first`
- Fix typos/grammar

Based on commit 5e51bdda.

We tenderized the wrong method! Object#try already had the yield option, just needed some tenderloving instance_eval to fit the bill

This reverts commit 70d6e16f, reversing
changes made to ea4db3bc.

Seems to be a code merge done by mistake.

Removed in 0228a73b, pull request #7310.

 will now return nil instead of raise a NoMethodError if the receiving object does not implement the method

[ci skip] closes #5790

This reverts commit 29a5aeaa.

Add support for try to just yield the object to a block if no method is to be called. Kind of like a tap_if_present.

  http://rdoc.rubyforge.org/RDoc/Parser/Ruby.html

Updated Object#try doc to match lifo's "no code change" revert c9b8a26b