add dra support

[luohua13]

Summary by CodeRabbit

• Documentation

  • Added a comprehensive guide for installing and verifying Alauda's NVIDIA DRA Driver for GPUs, covering prerequisites, configuration, installation steps, verification commands, and a validated workload example.

• Chores

  • Updated project ignore rules to exclude IDE-specific metadata.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

Walkthrough

Adds an ignore rule for the .idea directory to the repository .gitignore and introduces a new comprehensive documentation guide for installing, configuring, verifying, and using the Alauda Build of the NVIDIA DRA Driver for GPUs.

Changes

Cohort / File(s)	Summary
Configuration /.gitignore	Added a rule to ignore the .idea directory (IDE-specific metadata).
Documentation docs/en/solutions/How_to_Use_Alauda_Build_of_NVIDIA_DRA_Driver_for_GPUs.md	New file (~291 lines added) containing DRA concepts, prerequisites, containerd/CDI configuration, Kubernetes component flags and snippets, cluster plugin download/upload steps, DRA driver installation, verification examples (ResourceSlice, ResourceClaimTemplate), and a validated GPU workload with commands and YAML to review for accuracy.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Poem

🐇 I found a dot in .idea, snug and neat,
I hid it from commits with a silent feat.
Then through GPU guides I hopped and read,
With yaml, pods, and nvidia-smi fed.
A carrot-coded cheer — the changes complete.

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title 'add dra support' directly relates to the main change: adding documentation for NVIDIA DRA Driver support with a new guide file.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 3

🤖 Fix all issues with AI agents

In `@docs/en/solutions/How_to_Use_Alauda_Build_of_NVIDIA_DRA_Driver_for_GPUs.md`:
- Around line 132-134: Replace the incomplete docs link under the "Uploading the
Cluster plugin" section by updating the href to the specific documentation page
for uploading cluster plugins (instead of the docs homepage); locate the
"Uploading the Cluster plugin" heading and change the URL currently set to
https://docs.alauda.io/ to the correct target (e.g., the specific Uploading
Cluster Plugins page URL) so the anchor points directly to the detailed
instructions for uploading cluster plugins.
- Around line 158-163: Add a language identifier to the fenced code block
showing kubectl output so syntax highlighting renders correctly; update the
opening triple backticks for the snippet containing
"nvidia-dra-driver-gpu-controller-675644bfb5-c2hq4 ..." to use `text` (or
`console`) instead of a plain fence, e.g., change the fence from ``` to ```text
to mark the block as kubectl/terminal output.
- Around line 18-24: Update the two documentation lines under the headings
"Installing Nvidia driver in your GPU node" and "Installing Nvidia Container
Runtime" by replacing the incorrect phrase "Prefer to" with the correct "Refer
to" and ensure the link text reads naturally (e.g., "Refer to the Nvidia
official website installation guide" and "Refer to the Nvidia Container Toolkit
installation guide") so the references are grammatically correct and clearer.

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@docs/en/solutions/How_to_Use_Alauda_Build_of_NVIDIA_DRA_Driver_for_GPUs.md`:
- Around line 83-93: The kube-apiserver manifest uses multiple --runtime-config
flags which override each other; change the kube-apiserver container command
(the kube-apiserver args) to pass a single --runtime-config flag with
comma-separated entries e.g.
--runtime-config=resource.k8s.io/v1beta1,resource.k8s.io/v1beta2 so both API
versions are enabled; update the command list under the kube-apiserver container
in the manifest accordingly.

♻️ Duplicate comments (3)

docs/en/solutions/How_to_Use_Alauda_Build_of_NVIDIA_DRA_Driver_for_GPUs.md (3)

27-33: Fix “Prefer to …” phrasing in doc references.

Use “Refer to …” for correct grammar in documentation links.

📝 Proposed fix

-Prefer to [Installation guide of Nvidia Official website](https://docs.nvidia.com/cuda/cuda-installation-guide-linux/)
+Refer to [Installation guide of Nvidia Official website](https://docs.nvidia.com/cuda/cuda-installation-guide-linux/)

-Prefer to [Installation guide of Nvidia Container Toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html)
+Refer to [Installation guide of Nvidia Container Toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html)

---

139-141: Point the “Uploading Cluster Plugins” link to the specific page.

The link currently goes to the docs homepage, which is ambiguous. Please link to the actual “Uploading Cluster Plugins” page.

---

165-168: Add a language identifier to the fenced output block.

Use text or console so linters and renderers format it correctly.

📝 Proposed fix

-    ```
+    ```text
     nvidia-dra-driver-gpu-controller-675644bfb5-c2hq4   1/1     Running   0              18h
     nvidia-dra-driver-gpu-kubelet-plugin-65fjt          2/2     Running   0              18h
     ```