feat(profiles): auto-inject GPU jupyterlab image for gpu: true profiles

[tylerpotts]

What

Closes #230

Profiles in jupyterhub.custom.profiles can now be marked gpu: true:

- slug: gpu
  display_name: "GPU Access"
  gpu: true
  access: yaml
  groups: [gpu-access]
  kubespawner_override:
    # no image needed — the -gpu image is injected automatically
    node_selector: {node.kubernetes.io/instance-type: g4dn.xlarge}
    extra_resource_limits: {nvidia.com/gpu: 1}

At hub startup, 01-spawner.py injects kubespawner_override.image with the chart-derived GPU image, so GPU profiles track pack updates exactly like CPU profiles instead of pinning a -gpu SHA in the deployer overlay.

How

• Both jupyterlab images are built by the same build-images.yaml run from the same commit, so they always share the same sha-<short> tag. The chart derives the GPU ref as <singleuser.image.name>-gpu:<singleuser.image.tag> via a new gpuJupyterlabImage helper, exposed through _CHART_DERIVED as gpu-image (same pattern as nebi-image).
• _resolve_gpu_profiles() in 01-spawner.py runs once at load: injects the image when none is set, leaves an explicit kubespawner_override.image alone, and always strips the gpu key before KubeSpawner sees it.
• Deployers can override the injected ref chart-wide via jupyterhub.custom.gpu-image.
• scripts/bump_image_tags.py already bumps singleuser.image.tag each release, so the derived GPU ref stays current with zero script changes.

Testing

• tests/unit/test_spawner_profiles.py: injection, explicit-image precedence, gpu-key stripping, empty-derived-image fallback, input non-mutation, load-time wiring (7 new tests).
• tests/unit/test_chart_derived.py: rendered _CHART_DERIVED contains the derived gpu-image; explicit override wins (2 new tests).
• Full unit suite: 137 passed. helm lint / helm template (incl. --set nebariapp.enabled=false as CI does) and ruff check config/ clean.

Docs

• docs/src/content/docs/server-profiles.md: new "The GPU image, without hardcoding a SHA" section.
• docs/src/content/docs/values-reference.md: gpu-image row.
• values.yaml: gpu-image placeholder + updated GPU profile example.

[github-actions]

Docs preview for feat/gpu-profile-image-230:
https://feat-gpu-profile-image-230.data-science-pack.pages.dev