Rich t kid/implement multi dictionary aggr

[Rich-T-kid]

Which issue does this PR close?

• Closes #.

Rationale for this change

What changes are included in this PR?

Are these changes tested?

Are there any user-facing changes?

[Rich-T-kid]

alot of minor things to clean up.
notably most of the logic here isn't too specific to dictionary arrays besides caching the values pointer. we need benchmarks to determine if this is worth approaching over a generic implementation like GroupValueRows

[Rich-T-kid]

going through this now

[Rich-T-kid]

Still iterating on the PR, these are loose ideas for optimizations (intern path)

use VecDeque instead of Vec

• This should avoid the O(n) that happens for Vec<T>.drain()

adjust the row-tuple cache to be configurable

• currently set to 10,000, once it reaches this size the elements that were written to remain there but are never read again.
• need to strike a balance between having a cache for per batch keys as well as avoiding ever growing memory in the worst case
• Stop inserting after sparse-cache cap, but keep looking up
• Mabey a heap of the top K elements? K can be set as through SessionConfig?

avoid current double allocation spent for option

• options that wrap non pointer types generally take up more space. Option takes up 16 bytes instead of just 8
• mabey use Option<NonZeroUsize> 🤔

bitpack key-tuples cache

• instead of using a slice of usize (u64) bit back the combination of tuples into one u64 or 128
• need to do more research on this.

[Rich-T-kid]

see context as to why this isnt being pursued further #21765 (comment)