In a data engineering project, the team is utilizing Auto Loader to ingest data from a JSON source into a Databricks environment. They observe that despite the JSON source containing a mix of data types including integers, booleans, and strings, Auto Loader is inferring all data as STRING. This has led to data processing inefficiencies and inaccuracies in downstream analytics. Considering the need for accurate data type inference to ensure data quality and processing efficiency, which of the following is the MOST LIKELY reason for this behavior and the BEST solution to resolve it? Choose one option.