NO-ISSUE: docs(website): add metrics monitoring

[hhk7734]

No description provided.

[Copilot]

Pull request overview

This PR expands the website operations documentation by adding a new Grafana-based metrics monitoring guide (with supporting dashboard screenshots) and organizing the Monitoring section in the sidebar. It also introduces a new “Resource Allocation” operations guide and updates a prerequisite link in the log-collection doc.

Changes:

• Add Metrics Monitoring documentation under Operations → Monitoring, including Grafana access steps and metric/dashboard explanations.
• Add new Resource Allocation guide covering GPU/RDMA requests/limits, node selection, and taints/tolerations.
• Add Monitoring category metadata and fix the log-collection prerequisites link.

Reviewed changes

Copilot reviewed 4 out of 11 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
website/docs/operations/resource-allocation.mdx	New guide for resource requests/limits, node selection, and tolerations.
website/docs/operations/monitoring/metrics/index.mdx	New metrics monitoring guide explaining Grafana access and dashboard metrics.
website/docs/operations/monitoring/metrics/heimdall.png	Screenshot asset referenced by metrics guide.
website/docs/operations/monitoring/metrics/gpu-total.png	Screenshot asset referenced by metrics guide.
website/docs/operations/monitoring/metrics/filters.png	Screenshot asset referenced by metrics guide.
website/docs/operations/monitoring/log-collection.mdx	Fix prerequisites link formatting/target.
website/docs/operations/monitoring/category.yaml	Adds Monitoring section category configuration for the sidebar.

Comments suppressed due to low confidence (1)

website/docs/operations/resource-allocation.mdx:211

• Same as above: the +++ AMD GPUs / +++ NVIDIA GPUs / trailing +++ markers are not standard MDX/Docusaurus formatting and will likely render as stray text. Please convert these to headings or a supported grouping component (Tabs/admonitions/details) for readability.

+++ AMD GPUs

**Node taint:**

```yaml
taints:
  - key: 'amd.com/gpu'
    effect: 'NoSchedule'

Pod toleration:

tolerations:
  - key: 'amd.com/gpu'
    operator: 'Exists'
    effect: 'NoSchedule'

+++ NVIDIA GPUs

Node taint:

taints:
  - key: 'nvidia.com/gpu'
    effect: 'NoSchedule'

Pod toleration:

tolerations:
  - key: 'nvidia.com/gpu'
    operator: 'Exists'
    effect: 'NoSchedule'

+++

</details>