[DocDB] Add tool to parse results for Jepsen append workload failures

[es1024]

Jira Link: DB-14792

Description

The Jepsen append workload has two operations:

• Append an integer to a key's value
• Read value of a key

When the Jepsen test fails, it's typically by identifying cycles corresponding to anomalies, and the
set of transactions and all of their operations for each anomaly are dumped in the results. These
results are often difficult to read:

• Read values are printed in full, including all the values appended by transactions well before the
  first transaction of the cycle. They're also printed one value per line, resulting in hundreds of
  lines of output per transaction.
• Transaction IDs are not known to the Jepsen client, and thus not printed. It is typically
  necessary to identify each transaction in YB logs by dumping batches prepared/written and matching
  it against the operations reported by Jepsen.
• Tests are performed against multi-node clusters, so logs for a single transaction will span
  multiple log files.

A tool to address the above issues should be added to help debug Jepsen issues.

Issue Type

kind/enhancement

Warning: Please confirm that this issue does not contain any sensitive information

• I confirm this issue does not contain any sensitive information.