Very strange for sentinel monitor operation system load

## Issue Description

Type: *feature request*

### Describe what happened (or what feature you want)

Since we should pay notice to operation system system load,
but in sre sence, we offer just pay notice to the system load of 1 core.
For example , in k8s we may use this metric instance:node_load1_per_cpu:ratio
For developers, we just not want to know which kind node our application runs on.

### Describe what you expected to happen
We just want a new system load pre core for our system rule

### How to reproduce it (as minimally and precisely as possible)

### Tell us your environment

### Anything else we need to know?

No files, tests, or implementation entry points are named. Start by locating Sentinel's system-load monitoring and system-rule handling, then determine how the requested per-core load metric should be obtained without depending on the node type. Done means a system rule can use per-core load consistently, with coverage for the new behavior.

Written by the indexing model from the issue text.