Provide self-descriptive text for `selectdb/doris.fe-ubuntu` and `selectdb/doris.be-ubuntu` on Docker Hub

- I noticed from https://github.com/apache/shardingsphere/issues/29052 that the direct way to bypass compiling Doris Docker Image is to use https://github.com/apache/doris-operator on kubernetes.
- But using kubernetes operator in unit tests is very taxing on the `4C`/`16gb` Github Runner device. If I want to write integration tests for doris, I need to first start a `k3s` cluster with docker via `testcontainers-java`, and then deploy doris's kubernetes operator on the `k3s` cluster, which almost certainly causes `out of memory` issues on the Github Runner device because shardingsphere also resident additional `hotspot vm` processes or `graalvm native image` processes on the Github Runner device.
- I noticed that https://github.com/apache/doris-operator/blob/release-24.0.0/helm-charts/doris/values.yaml is using `selectdb/doris.fe-ubuntu` and `selectdb/doris.be-ubuntu`. But neither of these two Docker Images has self-descriptive text.
- 
- 
- It would be nice if these two Docker Images have self-descriptive text to describe how to use them. This would avoid dynamically downloading binaries and generating Docker Images in the context of `testcontainers-java`, and also avoid starting a `k3s` cluster.