Pruning of floating point Parquet columns is incorrect when `NaN` is present

### Describe the bug

This was mentioned in https://github.com/apache/datafusion/issues/15742#issuecomment-2815595171 and discussed in detail in https://github.com/apache/parquet-format/pull/221, but datafusion is over-aggressive in pruning floating point columns. The issue appears with predicates of the form `x [gt|lt] literal`. Consider a column consisting of `[1.0, 0.0, -1.0, NaN, -2.0]`, the max will be 1 and the min -2. A query like `select * from ... where x > 2` will return no rows because no chunk exists where `max > 2`.

### To Reproduce

```sql
> select * from 'parquet-testing/data/float16_nonzeros_and_nans.parquet' where x > arrow_cast(2.0, 'Float16');
+---+
| x |
+---+
+---+
0 row(s) fetched.
```

### Expected behavior

The above query should return a single row containing `NaN`.

### Additional context

The Parquet community is considering changes to allow for `NaN` in statistics, with the currently favored approach being adding a new `ColumnOrder` to the specification. This will correct the issue above, but datafusion will need to check the `ColumnOrder` to know whether or not floating point statistics can be trusted.

Also note that if/when https://github.com/apache/parquet-format/pull/221 is merged, other predicates such as `isnan(x)` might be candidates for pruning, but that is an optimization.

Start by running the query against parquet-testing/data/float16_nonzeros_and_nans.parquet and confirm that the NaN row is incorrectly pruned. Trace floating-point statistics pruning and how Parquet ColumnOrder is handled, using the linked parquet-format discussion for context. Done means the query returns the NaN row without over-aggressive pruning, with regression coverage for the reproduction.

Written by the indexing model from the issue text.