[Improvement] Coordinate support mixed storage shuffle servers

### Code of Conduct

- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)

### Search before asking

- [X] I have searched in the [issues](https://github.com/apache/incubator-uniffle/issues?q=is%3Aissue) and found no similar issues.

### What would you like to be improved?

After the https://github.com/apache/incubator-uniffle/issues/405 is finished, shuffle server would report its capability to coordinator, such as
- disk quota
- disk iops(by HDD/SSD)
- disk number
- server thread number
etc.

The coordinator should support such mixed deployments, and assign workload based on shuffle worker's capability. In this way, one uniffle cluster could scale up to handle more traffic with commodity servers and also support high priority apps with specialized hardware, such as server with SSD disks.

### How should we improve?

- implement new assignment strategy to support:
- more affinity settings: such as prefer, required semantics
- load awareness assignments
- necessary changes to modify corresponding RPC interfaces.

### Are you willing to submit PR?

- [ ] Yes I am willing to submit a PR!

Start by reviewing issue #405, then trace the coordinator's current assignment strategy and the RPC interfaces mentioned in the issue. Done means supporting mixed shuffle-server capabilities with preference or requirement affinity and load-aware workload assignment.

Written by the indexing model from the issue text.