JIT-Compile DataFusion Expressions to create RecordBatches

[Dandandan]

Is your feature request related to a problem or challenge? Please describe what you are trying to do.
We should be able to compile.

The benefit of this is that we can speed up complex / nested expressions by avoiding unnecesarry allocations

Describe the solution you'd like

We should be able to take in a collection of a RecordBatch / named Arrays and compile an expression like (a + b)/ 2 to a loop that results in a new Array.

fn compile(schema: SchemaRef, expr: Expr) ->  CompiledFunction {

}

The loop itself also must be included in the to-be compiled expression, to remove call overhead and allow for possible use of SIMD instructionsinstructions, either explicitly by instrumenting cranelift enough or through auto-vectorization.

Describe alternatives you've considered
n/a

Additional context

[alamb]

datafusion PhysicalExpr and arrow-rs library currently evaluate expressions by "materializing intermediate results" -- for example (a + b) + c results in first evaluating (a+b) to a temporary location and then adding c to form the final result.

Note however, there is a tradeoff here between the speed gained using the LLVM optimized vectorized kernels in arrow-rs and cranelift generated JIT expressions where JIT may not actually be faster. I think this is what @Dandandan is referring to when he says "allow for possible use of SIMD instructions, either explicitly by instrumenting cranelift enough or through auto-vectorization."

[alamb]

Another example can be found in these slides from this presentation