Tracking: improve aggreagation fuzzer

### Is your feature request related to a problem or challenge?

I found aggregation fuzzer are still hard to use when I act as an user currently.
Some points I noticed can be improved:

### 1. Aggregation function generation
Now it is still hard/possible to generate sql having multiple aggregated functions like:
```text
SELECT count(string), sum(number) FROM xxx GROUP BY xxx;
```

It is due to `aggregate_function` and `aggregate_arguments` are set respectively. if we set `mutiple aggregate_functions`, columns in the respectly set `aggregate_arguments` may be not supported in some of the `aggregate_functions`.
For example,
```
If we set:
aggregate_arguments: [string, primitive]
aggregate_function: [sum, count]

Then wrong generated sql possible to be generated:
SELECT sum(string) FROM t GROUP BY xxx
```

### 2. Group by columns ordering
Noticed by @jayzhan211 , it is hard to control the ordering of group by columns right now.
https://github.com/apache/datafusion/pull/15851#issuecomment-2831689650

### Describe the solution you'd like

_No response_

### Describe alternatives you've considered

_No response_

### Additional context

_No response_

The issue names the aggregation fuzzer but no file or test entry point. Start by locating the aggregate_function, aggregate_arguments, and GROUP BY generation code; then add coverage for compatible aggregate arguments and controllable GROUP BY column ordering, including the examples described in the issue.

Written by the indexing model from the issue text.