Investigate detection of Literal and TemplateLiteral from a bottom up approach

We are currently using multiple AST nodes (CallExpression BinaryExpression ObjectExpression etc) in order to find Literal and TemplateLiteral nodes and then perform the linting on them in a best case scenario. This means that we only detect the Literal and TemplateLiteral nodes in the cases we add handling for.

Let's spike moving away from this approach and instead add handling for Literal and TemplateLiteral and then work our way up from there.

The benefit of this is a greater detection that will remain scalable when new expressions are added to javascript -> such as the arrow expression.