Sort out tests in `aggregate.slt`

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

I found tests in `aggregate.slt` a bit messy and the file is too big(6088 rows)... when I try add more tests of `median` for #13681

Maybe we should split and sort out such tests to make it clearer.

### Describe the solution you'd like

I think https://github.com/apache/datafusion/pull/12525 is a good example.

And maybe we can do something like it. We can split the tests into multiple test files, and each file only contains testcases for a special aggregation function:

- Current
```
aggregate.slt
```
- After sorting out
```
aggregate
init.slt.part
min_max.slt
median.slt
...
```

### Describe alternatives you've considered

_No response_

### Additional context

_No response_

Start by reviewing the 6,088-line aggregate.slt file and the structure used in PR #12525. Split the aggregation tests into focused files such as init.slt.part, min_max.slt, and median.slt, while preserving the existing test coverage and ensuring the reorganized suite still runs successfully.

Written by the indexing model from the issue text.