Add unit test for gpuPodSpecFilter container resource-list branch

[abrarshivani]

Description

gpuPodSpecFilter decides whether a pod is a GPU pod via two branches:

1. container resource-list — a container requesting nvidia.com/gpu* or nvidia.com/mig-* in its limits/requests (gated on the pod being Running/Pending), and
2. DRA / ResourceClaims — PodHasNVIDIAGPUClaim(...).

The existing TestGPUPodSpecFilterResourceClaims covers only branch (2). Per the coverage report, branch (1) — the nvidia.com/gpu / nvidia.com/mig- prefix match and the container-loop return true — was uncovered.

This PR adds TestGPUPodSpecFilterResourceList to cmd/gpu-operator/main_test.go covering branch (1):

• nvidia.com/gpu in a container's limits and in requests
• the nvidia.com/mig- prefix
• a multi-container pod where only a later container requests a GPU
• a pod with only cpu/memory resources (not a GPU pod)
• the phase gate (a Succeeded pod with a GPU resource is ignored)

The pods carry no ResourceClaims, so the DRA branch resolves to false and only the resource-list branch is exercised. This complements the existing DRA test rather than duplicating it, bringing gpuPodSpecFilter to 100% statement coverage.

Tests only — no production changes.

Checklist

• No secrets, sensitive information, or unrelated changes
• Lint checks passing (make lint)
• Generated assets in-sync (make validate-generated-assets)
• Go mod artifacts in-sync (make validate-modules)
• Test cases are added for new code paths

Testing

go test ./cmd/gpu-operator/... -covermode=count    # gpuPodSpecFilter: 100.0%
golangci-lint run ./cmd/gpu-operator/...           # 0 issues (GOOS=linux)

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[tariq1890]

Can you rebase this PR?

[abrarshivani]

@tariq1890 I've rebased the branch. Can you please take another look?