What is the ideal breakdown of the testing pyramid for Gatsby?

I have been pondering over this issue for a while. The question I have is whether or not it is more useful to change the traditional testing pyramid structure,

• 70% unit tests
• 20% integration tests
• 10% end-2-end testing

to follow a pyramid that is more like,

• 70% end-2-end testing
• 20% unit testing
• 10% integration tests

For a website made in Gatsby.

The reasoning for this is because I am not seeing the value of unit testing components that are made purely for visuals but since I am a junior developer, I would like to hear your opinion on this.

Thanks.