BulkIngester: Add pipeline to index and create operations

[renatogm24]

Currently, when using the BulkIngester in the Elasticsearch Java client, the pipeline for individual index operations is not set up. Instead, the default pipeline on the index is used for all operations. This can limit flexibility and control when different documents in the same index require different processing.

For example, consider a scenario where we have multiple documents in the same index, but each document requires a different pipeline for processing. In the current implementation, all documents would be processed using the default pipeline on the index, which may not be suitable for all documents.

This issue can be addressed by allowing the pipeline to be set up for each individual index operation. This would involve copying the pipeline property from the IndexOperation to the IndexOperation.Builder for each operation.

I have created a pull request that introduces these changes: PR #587. Please review the PR and let me know if you have any questions or need further information.

[swallez]

Fixed in PR #587 and backported to the 8.8 branch in PR #594