Spark 3.5: Support camel case session configs and options

[aokolnychyi]

This PR adds support for parsing camel case session configs and options. Our keys contain - but all built-in Spark configs follow the camel case style. Apart from the inconsistency, setting Iceberg SQL configs requires escaping the keys, which makes it confusing for the users.

The following statement throws an error and requires the user to manually deal with -.

SET spark.sql.iceberg.planning.preserve-data-grouping = false

After this change, users will be able to set SQL properties and options either using - or camel case.

SET spark.sql.iceberg.planning.preserveDataGrouping = false
SET `spark.sql.iceberg.planning.preserve-data-grouping` = false

We already have a precedent for this in SparkWriteOptions.

public static final String MERGE_SCHEMA = "merge-schema";
public static final String SPARK_MERGE_SCHEMA = "mergeSchema";