Size Mapping V2 supports 'Identical Ad Units'

[Fawke]

Type of change

• Feature

Description of change

This PR introduces support for 'Identical Ad Unit' declaration in Size Mapping V2. An Identical Ad Unit is an ad unit with the same code as one of the existing ad units but different mediaTypes object.

Example of Identical Ad Units:

[
  {
    code: 'ad-code-1',
    mediaTypes: {
      banner: {
        sizeConfig: [
          { minViewPort: [0, 0], sizes: [] },
          { minViewPort: [1000, 0], sizes: [[1000, 300], [1000, 90], [970, 250], [970, 90], [728, 90]] },
        ],
      },
    },
    bids: [{
      bidder: 'bidderA',
      params: {
        placementId: 123
      }
    }]
  }, {
    code: 'ad-code-1',
    mediaTypes: {
      banner: {
        sizeConfig: [
          { minViewPort: [0, 0], sizes: [] },
          { minViewPort: [320, 0], sizes: [[320, 100], [320, 50], [300, 50]] },
          { minViewPort: [600, 0], sizes: [] },
        ],
      },
    },
    bids: [{
      bidder: 'bidderA',
      params: {
        placementId: 456
      }
    }]
  },
]

The example above has two identical ad units with the same code ad-code-1 but with different values for mediaTypes object.

The PR also improves the log messages generated by Size Mapping V2, tries to make it more intuitive and easy to understand.

Resolves #5044

[Fawke]

This PR is ready for review. Only thing that I'd have liked to do was remove some ambiguity from the log messages.

For example:

If we have two Identical ad units (sharing the same code), from the log messages it is kind of hard to determine which ad unit got disabled, or which ad unit got an error, since ad unit code is no longer a unique identifier.

In the screenshot above, we have two identical ad units sharing the same code ad-code-1, in which after sizeMapping filtration, the second instance of ad-code-1 gets disabled. But looking at the log messages, there's currently no way to tell was it the first instance or the second instance.

Once this PR is merged, I'll create a separate github issue to keep track of this behaviour and provide a fix in a separate PR.

[robertrmartinez]

Nice, looks pretty good!

I am always a big fan of improving our logging.