🚀[Feature Request]: GPU and CPU memory both should be queryable via ClusterScope

### A clear and concise description of what you want to happen.

This method implicitly assumes that clients are only interested in GPU-memory. Let's add support for both CPU and GPU memory and rename existing method(s).

```
def get_mem_per_node_MB(self) -> int:
"""Return the lowest amount of mem configured across all nodes in the cluster. Returns 0 if not a Slurm cluster.
```

### A clear and concise description of any alternative solutions or features you've considered, if any.

_No response_

### Additional context

_No response_