Update TrustyAI and jupyter-bokeh to 4.x series after upstream resolution

[coderabbitai]

Background

The TrustyAI workbench image was previously unable to upgrade jupyter-bokeh from 3.x to 4.x due to a hard dependency constraint in the upstream TrustyAI package. This has now been resolved!

Upstream Resolution ✅

• Issue: Consider updating jupyter-bokeh dependency from ~=3.0.5 to support 4.x series trustyai-explainability/trustyai-explainability-python#222
• Status: Implemented - TrustyAI now supports jupyter-bokeh 4.x series
• Impact: TrustyAI workbench can now align with other notebook images

Required Updates

1. Update TrustyAI Package Version

• Current: trustyai = "~=0.6.1"
• Target: Update to latest version that supports jupyter-bokeh 4.x
• Files:
  • jupyter/trustyai/ubi9-python-3.11/Pipfile
  • jupyter/trustyai/ubi9-python-3.12/Pipfile (if exists)

2. Update jupyter-bokeh Version

• Current: jupyter-bokeh = "~=3.0.5"
• Target: jupyter-bokeh = "~=4.0.5" (align with other images)
• Files: Same as above

3. Update Comment/Documentation

• Update or remove the compatibility comment about jupyter-bokeh pinning
• Align with the dependency pinning strategy used in other images

Acceptance Criteria

• TrustyAI package updated to version supporting jupyter-bokeh 4.x
• jupyter-bokeh updated to ~=4.0.5 in TrustyAI images
• Pipenv lock files regenerated successfully
• CI builds pass without dependency conflicts
• TrustyAI functionality verified in test notebooks

Context & References

• Related PR: RHOAIENG-21842: bump JupyterLab-related dependencies in all Python 3.11 and 3.12 Pipfiles #1218 (JupyterLab dependencies update)
• Original Discussion: RHOAIENG-21842: bump JupyterLab-related dependencies in all Python 3.11 and 3.12 Pipfiles #1218 (comment)
• Dependency Conflict: Previously blocked by trustyai 0.6.1 requiring jupyter-bokeh~=3.0.5
• Resolution: Upstream TrustyAI now supports jupyter-bokeh 4.x

Implementation Notes

This update will complete the dependency alignment effort across all notebook images, ensuring consistent Jupyter ecosystem versions throughout the workbench.

After this update, all notebook images will be using jupyter-bokeh 4.x series! 🎉