Fine grained handling of metrics scraping failures

[ahg-g]

What would you like to be added:

Currently model server metrics are not updated if we fail to find any of the metrics (e.g., if lora info doesn't exist, we don't update the kv-cache and queue length metrics). We should not have an all-or-nothing behavior and update whatever metrics we find, but we should still log that not all metrics where updated.

Why is this needed:

Not all servers support lora info now.

[ahg-g]

/assign @BenjaminBraunDev

[k8s-ci-robot]

@ahg-g: GitHub didn't allow me to assign the following users: BenjaminBraunDev.

Note that only kubernetes-sigs members with read permissions, repo collaborators and people who have commented on this issue/PR can be assigned. Additionally, issues/PRs can only have 10 assignees at the same time.
For more information please see the contributor guide

In response to this:

/assign @BenjaminBraunDev

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.

[BenjaminBraunDev]

This should allow us to do this cleanly: #503