Parameters are arriving to source/sink lowercase.
maxOffsetsPerTrigger parameter as an example:
KafkaSourceProvideruses caseInsensitiveParams which converts keys to lowercaseKafkaMicroBatchStreamusesCaseInsensitiveStringMapwhere get operation uses lowercase conversionKafkaSourceusesCaseInsensitiveMapwhere get operation uses lowercase conversion
In the last case CaseInsensitiveMap extends Map and as said it provides lowercase key lookup but in the [interface](https://github.com/apache/spark/blob/3e4