adding support for partitioned s3 source

[abhishekd0907]

Background

S3-SQS source doesn't support reading partition columns from the S3 bucket. As a result, the dataset formed using S3-SQS source doesn't contain the partition columns leading to issue #2

How this PR Handles the Problem

With the new changes, the user can specify partition columns in the schema with isPartitioned set to true in column metadata.

Example:

import org.apache.spark.sql.types._
import org.apache.spark.sql.types.MetadataBuilder

val metaData = (new MetadataBuilder).putString("isPartitioned", "true").build()

val partitionedSchema = new StructType().add(StructField("col1", IntegerType, true, metaData))

Also, the user needs to specify the basePath in options if the schema contains partition columns. Specifying partitioned
columns without specifying the basePath will throw an error.

Example:

 `s3://bucket/basedDir/part1=10/part2=20/file.json` will have basePath as `s3://bucket/basedDir/`

[itsvikramagr]

is this constraint which this module is adding or is it normally expected that the partition column will have isPartitioned in their metadata.