Deduplicate uncorrelated scalar subqueries

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

Per https://github.com/apache/datafusion/pull/21240#discussion_r3065275637 , if the same subquery (identical SQL text) appears multiple times in a query, it will be planned and evaluated multiple times. In principle it should be fine to evaluate it only once, provided the subquery doesn't contain any volatile functions (e.g., `random()`).

### Describe the solution you'd like

_No response_

### Describe alternatives you've considered

_No response_

### Additional context

_No response_

Start with the discussion in apache/datafusion#21240, which identifies the repeated planning and evaluation of identical scalar subqueries. Trace the query-planning path for uncorrelated scalar subqueries and determine how volatile functions such as random() are identified. Done means identical non-volatile subqueries are evaluated once while volatile subqueries retain their current behavior.

Written by the indexing model from the issue text.