Update interaction with failure message on agent execution failure

[pyek-bot]

Description

When an agent runs, it creates a new interaction. On successful execution, this interaction is updated with the final response. However, on agent failures, the interaction never gets updated. Therefore, this PR addresses that gap.

Example:
Before in case of agent failures, response field is never populated

{
    "memory_id": "Uw_ufZkBE5GN_tuwO2KT",
    "message_id": "Wg_yfZkBE5GN_tuwuGI9",
    "create_time": "2025-09-24T22:58:02.939633Z",
    "input": "give me index mapping of any index",
}

After

{
    "memory_id": "Uw_ufZkBE5GN_tuwO2KT",
    "message_id": "Wg_yfZkBE5GN_tuwuGI9",
    "create_time": "2025-09-24T22:58:02.939633Z",
    "input": "give me index mapping of any index",
    "response": "Agent execution failed: failed ot run"
}

Related Issues

Resolves #4197

Check List

• New functionality includes testing.
• New functionality has been documented.
• API changes companion pull request created.
• Commits are signed per the DCO using --signoff.
• Public documentation issue/PR created.

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.

[codecov]

Codecov Report

❌ Patch coverage is 82.85714% with 6 lines in your changes missing coverage. Please review.
✅ Project coverage is 81.74%. Comparing base (4864f66) to head (5b8aa24).
⚠️ Report is 2 commits behind head on main.

Files with missing lines	Patch %	Lines
...ch/ml/engine/algorithms/agent/MLAgentExecutor.java	82.85%	3 Missing and 3 partials ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##               main    #4198      +/-   ##
============================================
+ Coverage     81.69%   81.74%   +0.04%     
- Complexity     9374     9384      +10     
============================================
  Files           796      796              
  Lines         40622    40652      +30     
  Branches       4515     4519       +4     
============================================
+ Hits          33186    33230      +44     
+ Misses         5490     5471      -19     
- Partials       1946     1951       +5     

Flag	Coverage Δ
ml-commons	81.74% <82.85%> (+0.04%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[pyek-bot]

CI failure is unrelated & flaky

130 tests completed, 1 failed, 11 skipped
Tests with failures:
 - org.opensearch.ml.rest.RestMLRAGSearchProcessorIT.testBM25WithBedrockConverseUsingLlmMessages

[ylwu-amzn]

Rebase with main branch ?

[pyek-bot]

Rebase with main branch ?

Rebased!

[mingshl]

This is a good start, but we just do a general error message now. later when we know the pattern of the errors. we better to do different error handling for different types of failures (validation, execution, timeout). That would be better for user to understand rather than throwing the raw exception message.

[pyek-bot]

Thanks @mingshl Agreed!

Another improvement I notice:

• A lot of the agent runners handle same logic in their individual classes
• This can be centralized my moving the changes to the AgentExecutor (like returning a standard result)

This PR is a start towards that as the error message handling is happening in a common place.