Review parameter type declarations

[jrfnl]

While looking at something different, we noticed a number of places where parameters did not have (non-scalar) type declarations, while they could have.

It would probably be a good idea to do a review of the complete codebase to add object-based/array type declarations wherever possible.

This would not be a breaking change as overloading methods not having a type declaration, means those are declared as mixed and mixed is contravariant with any type, so this would not lead to signature mismatch errors.

[jrfnl]

Currently having a prelim look at this and I am now doubting we can actually do this in a minor as certain code would either:

• start throwing a TypeError/Exception where previously it would silently accept the incorrect type and somehow handle it; or
• the code would previously throw a different type of exception ( WpOrg\Requests\Exception\InvalidArgument most typically), so the change of which exception is being thrown could be seen as a breaking change as WpOrg\Requests\Exception\InvalidArgument does not extend TypeError.

With that in mind, I'm moving this to the 3.0 milestone.