vLLM inference servers container image fixes

[seans3]

• Updates the container image tag for vLLM inference server to the most recent (v0.11.0).
• Adds LD_LIBRARY_PATH environment variables to make the server more robust. This environment variable can be missing from some container images.
  • Ref: [Bug]: Image v0.9.0 Fails to Initialize on GCP instance Due to Undetected Platform vllm-project/vllm#18859
• Adds vLLM parameters to ensure no out-of-memory issues (for this model).

[seans3]

/assign @janetkuo

[janetkuo]

Is the information here meaningful to users? Is LD_LIBRARY_PATH only needed for certain version of vLLM? If so, that kind of information is good to capture here.

[seans3]

Good point. I've added the versions of the image that this is needed for. I've kept a reference to the issue, in case users need to do more research. Please let me know what you think.

[janetkuo]

/lgtm

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: janetkuo, seans3

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [janetkuo]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment