Add individual doc retries to BulkVectorDataSet runner

[OVI3D0]

Description

Adds simple/detailed stats methods to the BulkVectorDataSet runner and also changes the retry logic.

Rather than retrying the entire bulk upon failure, this PR adds logic to parse through failed documents and retry only those failed ones rather than the entire bulk.

2025-11-13 01:32:47,15 ActorAddr-(T|:41429)/PID:616400 osbenchmark.worker_coordinator.runner INFO Retrying 54 structured docs, indices=[0, 1, 2, 5, 8, 9, 11, 12, 13, 15]

Ingested documents using a 1M dataset with these changes:

[ec2-user@ip-172-31-0-197 ~]$ ~/.local/bin/awscurl --service aoss --region "$REGION"   --access_key "$AWS_ACCESS_KEY_ID"   --secret_key "$AWS_SECRET_ACCESS_KEY"   --security_token "$AWS_SESSION_TOKEN"   -X GET "$AOSS_ENDPOINT/${INDEX}/_count?pretty"
{
  "count" : 999990,
  "_shards" : {
    "total" : 0,
    "successful" : 0,
    "skipped" : 0,
    "failed" : 0
  }
}

vs without the changes:

[ec2-user@ip-172-31-0-197 ~]$ ~/.local/bin/awscurl --service aoss --region "$REGION"   --access_key "$AWS_ACCESS_KEY_ID"   --secret_key "$AWS_SECRET_ACCESS_KEY"   --security_token "$AWS_SESSION_TOKEN"   -X GET "$AOSS_ENDPOINT/${INDEX}/_count?pretty"
{
  "count" : 525869,
  "_shards" : {
    "total" : 0,
    "successful" : 0,
    "skipped" : 0,
    "failed" : 0
  }
}

Issues Resolved

#979

Testing

• New functionality includes testing

Running ingestion of 1M documents on AOSS vectorsearch collection

---

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.