Add params_to_tune for ma models

[Mr-Geekman]

🚀 Feature Request

Add params_to_tune for ma models.

Proposal

Suggested grid for `SeasonalMovingAverageModel:

{
    "window": IntUniform(low=1, high=10, step=1),
}

Suggested grid for MovingAverageModel:

{
    "window": IntUniform(low=1, high=10, step=1),
}

Suggested grid for NaiveModel:

{}

Suggested grid for DeadlineMovingAverageModel:

{
    "window": IntUniform(low=1, high=10, step=1),
}

To discuss:

• Can we do something with seasonality parameter in SeasonalMovingAverageModel? The problem is that it depends on the data. For example, we can just keep the value that was put by the user and don't change it.
• Can we do something with lag parameter in NaiveModel? The problem is that it depends on the type of pipeline that we use, the horizon and the data. We can't make lag smaller than horizon for Pipeline. We can make it bigger but does it make sense?
• Should we tune seasonality of DeadlineMovingAverageModel?
  • Currently decided not to tune it.

Test cases

Look at #1184.

Additional context

No response

[Mr-Geekman]

Closed by #1208.