Usually, the AZ Data Lake is used as a storage which is much cheaper compared to other tools and you can manage the jobs side via Databricks notebooks (we are talking here about running jobs with high-performance and in-memory clusters).
Another detail Databricks has more languages options and it is dedicated to many backgrounds working with data.
If you are dealing with complex batch process and you have different type of developers who will work on your project you may go for both( a Azure Data Lake + Databricks architecture). Otherwise you can use Azure Data Lake.
Here are some links :
https://www.element61.be/en/resource/when-use-azure-synapse-analytics-andor-azure-databricks
https://visualbi.com/blogs/microsoft/azure/etl-azure-databricks-vs-data-lake-analytics/