fix(llmobs): fix assessment argument description for custom evals (#15032)

## Description
Corrects the docstring description of the `assessment` argument in
`submit_evaluation()`. There is no functional change but the docstring
description was incorrect.
<!-- Provide an overview of the change and motivation for the change -->

## Testing

<!-- Describe your testing strategy or note what tests are included -->

## Risks

<!-- Note any risks associated with this change, or "None" if no risks
-->

## Additional Notes

<!-- Any other information that would be helpful for reviewers -->

Author: Yun-Kim

ddtrace/llmobs/_llmobs.py

@@ -1630,7 +1630,7 @@ def submit_evaluation(
                                     If not set, the current time will be used.
         :param dict metadata: A JSON serializable dictionary of key-value metadata pairs relevant to the
                                 evaluation metric.
-        :param str assessment: An assessment of the validity of this evaluation. Must be either "pass" or "fail".
+        :param str assessment: An assessment of this evaluation. Must be either "pass" or "fail".
         :param str reasoning: An explanation of the evaluation result.
         """
         if span_context is not None:

releasenotes/notes/fix-llmobs-evals-assessment-98f964a9f249913e.yaml

@@ -0,0 +1,5 @@
+---
+fixes:
+  - |
+    LLM Observability: Corrected the description of the ``assessment`` argument in ``submit_evaluation()``.
+     ``assessment`` now refers to whether the evaluation itself passes or fails according to your application, rather than the validity of the evaluation result.

tests/llmobs/test_llmobs_service.py

@@ -1879,7 +1879,7 @@ def test_submit_evaluation_enqueues_writer_with_assessment(llmobs, mock_llmobs_e
         tags={"foo": "bar", "bee": "baz", "ml_app": "ml_app_override"},
         ml_app="ml_app_override",
         metadata={"foo": ["bar", "baz"]},
-        assessment="pass",
+        assessment="fail",
     )
     mock_llmobs_eval_metric_writer.enqueue.assert_called_with(
         _expected_llmobs_eval_metric_event(
@@ -1891,7 +1891,7 @@ def test_submit_evaluation_enqueues_writer_with_assessment(llmobs, mock_llmobs_e
             categorical_value="high",
             tags=["ddtrace.version:{}".format(ddtrace.__version__), "ml_app:ml_app_override", "foo:bar", "bee:baz"],
             metadata={"foo": ["bar", "baz"]},
-            assessment="pass",
+            assessment="fail",
         )
     )
     mock_llmobs_eval_metric_writer.reset()