Remove job related metrics/dashboards

[azreenz]

Remove metrics with high cardinality labels to reduce prometheus backend explosion

[Copilot]

Pull request overview

This PR removes per-job (high-cardinality) Prometheus metrics and the Grafana dashboards/panels that depended on them, to reduce Prometheus/Azure Monitor backend series growth and associated operational risk.

Changes:

• Removed the per-job squeue_job_nodes_allocated metric from the squeue collector and updated unit tests accordingly.
• Removed the per-job sacct_failed_jobs metric from the sacct collector and updated unit tests accordingly.
• Deleted the “Failed Jobs” dashboard and removed job/failed-job panels from the “Extra Details” dashboard; updated dashboard navigation text and README exported-metrics documentation.

Criterion	Max Points	Points Awarded	Notes
PR Description Accuracy	20	20	Matches changes (removal of high-cardinality metrics/dashboards).
PR Atomicity	20	20	Single cohesive change: remove job-level metrics and dependent dashboards/tests/docs.
Logical Implementation	10	10	Straightforward removals; remaining collectors align with updated behavior.
Regression Risk	10	10	Change is intentionally breaking for removed metrics/dashboards and is described.
Exception Handling	10	10	No degradation vs existing error handling patterns.
Code Comments	10	10	Docstrings updated to reflect the new single-metric outputs.
Repetitive Code	10	10	No new duplication introduced.
Spelling	5	5	No spelling issues found in modified text.
Logging Quality	5	5	No negative changes to logging clarity/levels.

FINAL SCORE: 100/100

RECOMMENDATION: MERGE
RATIONALE: The PR cleanly removes the targeted high-cardinality metrics and updates tests/docs/dashboards consistently.
BLOCKERS: None

Reviewed changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
README.md	Removes documentation for the deleted job-level metrics.
azure-slurm-exporter/exporter/squeue.py	Removes per-job node allocation metric; keeps partition/state job counts.
azure-slurm-exporter/exporter/sacct.py	Removes per-job failed-jobs gauge; keeps terminal-jobs counter.
azure-slurm-exporter/test/test_squeue.py	Updates expectations to the single remaining squeue metric.
azure-slurm-exporter/test/test_sacct.py	Updates expectations to the single remaining sacct metric.
azure-slurm-exporter/dashboards/library-panels/azslurm-dashboard-buttons.json	Updates button label text for the details dashboard.
azure-slurm-exporter/dashboards/failed-jobs.json	Deletes the failed-jobs dashboard.
azure-slurm-exporter/dashboards/azslurm-details.json	Removes job/failed-job panels and updates dashboard description/layout.

[azreenz]

new extra details dashbaord thast doesnt contain any job specific panels