[Benchmark] Add E2E Benchmark Support for KV and LOG Workloads

### Search before asking

- [x] I searched in the [issues](https://github.com/alibaba/fluss/issues) and found nothing similar.

### Description

Currently, we lack an end-to-end (E2E) benchmark tool to efficiently generate performance reports for various workloads. This makes it difficult to evaluate the performance impact of every code change.

We should introduce a convenient E2E benchmark suite. It can be used to build a daily performance regression test job and also run in a local environment during feature development.

### Willingness to contribute

- [ ] I'm willing to submit a PR!

Start by examining the repository's existing benchmark and end-to-end test infrastructure, then identify how KV and LOG workloads can be generated and measured. Done means a convenient suite produces performance reports for both workloads and can support daily regression jobs as well as local feature development.

Written by the indexing model from the issue text.