fix: MLFlow E2E Example Notebook (5513)

[aviruthen]

Description

The notebook at v3-examples/ml-ops-examples/v3-mlflow-train-inference-e2e-example.ipynb already exists but has several bugs related to MLflow 3.x API compatibility, incorrect V3 SDK Session usage, and inconsistency with existing V3 notebook patterns. Key issues: (1) Session.boto_region_name used as class attribute instead of instance attribute, (2) MLflow 3.x removed RegisteredModel.latest_versions — must use client.search_model_versions() instead, (3) client.get_model_version_download_uri() doesn't exist in MLflow 3.x — use model version's source attribute, (4) Endpoint invocation uses raw boto3 instead of core_endpoint.invoke() pattern from other V3 example notebooks, (5) The notebook should follow consistent patterns with the existing train-inference-e2e-example.ipynb.

Related Issue

Related issue: 5513

Changes Made

• v3-examples/ml-ops-examples/v3-mlflow-train-inference-e2e-example.ipynb

AI-Generated PR

This PR was automatically generated by the PySDK Issue Agent.

• Confidence score: 85%
• Classification: bug
• SDK version target: V3

Merge Checklist

• Changes are backward compatible
• Commit message follows prefix: description format
• Unit tests added/updated
• Integration tests added (if applicable)
• Documentation updated (if applicable)

[sagemaker-bot]

🤖 AI Code Review

This PR fixes several real bugs in the MLflow E2E example notebook: replacing deprecated MLflow 3.x APIs (latest_versions, get_model_version_download_uri) and switching from raw boto3 endpoint invocation to the V3 SDK's core_endpoint.invoke() pattern. The changes are correct and align with V3 SDK conventions. A minor concern about the MLflow search API parameter name.