Added logic to include partition metadata for no-copy

[subkanthi]

closes: #46

[subkanthi]

 insert btc.transactions_no_copy --no-copy --force-no-copy --s3-region=us-east-2 --s3-no-sign-request -p s3://aws-public-blockchain/v1.0/btc/transactions/date=2025-01-01/part-00000-33e8d075-2099-409b-a806-68dd17217d39-c000.snappy.parquet --partition='[{"column": "block_timestamp", "transform": "month"}]'

[subkanthi]

To set the partitionKey to a datafile, we need the partition values

PartitionKey partitionKey = new PartitionKey(table.spec(), table.schema());
// Assuming your S3 path already reflects the partitioning, you'd extract the partition values from the path
// For example, if your path is s3://your-bucket/path/to/your/datafile_hour=10.parquet, you'd set the hour for the partitionKey.
// However, for DataFile, the partitionKey should represent the partition values within the file itself.
// You would then build the PartitionKey from the record's values to match the partition spec.

// If you need to manually set the partition values for the DataFile:
partitionKey.set(0, hour_value); // Set the value for the first partition field (e.g., hour)

[shyiko]

Closing in favor of logic included in 0.5.0 that infers key from the metadata instead of data. https://github.com/Altinity/ice/pull/51/files#diff-efe5f830dfd30841f29c50a9f843a6c295aafb4bfd3d60202bb22f8680272686R559 is also not a valid way to determine partitioning key unless data is guaranteed to be pre-partitioned.