[Python][RDF] Add support for std::vector and std::array in numba jitted functions

[siliataider]

This Pull request:

Changes or fixes:

This PR adds support for using std::vector<T> and std::array<T> as input and return types in functions declared with ROOT.Numba.Declare used in RDataFrame .Define operations.

This makes it possible to define and use functions like:

@ROOT.Numba.Declare(["std::array<int, 3>"], "std::vector<int>")
def square_vec(x):
    return x * x

rdf = ROOT.RDataFrame(3) \
        .Define("xvec", "std::array{(int)rdfentry_, (int)rdfentry_ + 1, (int)rdfentry_ + 2}") \
        .Define("xvec_square", "Numba::square_vec(xvec)")

The implementation reuses the logic for handling RVecs.

Checklist:

• tested changes locally
• updated the docs (if necessary)

[github-actions]

Test Results

    21 files      21 suites   3d 12h 9m 3s ⏱️
 3 223 tests  3 220 ✅ 0 💤 3 ❌
65 956 runs  65 952 ✅ 1 💤 3 ❌

For more details on these failures, see this check.

Results for commit 662ee98.

♻️ This comment has been updated with latest results.

[vepadulano]

Thank you!

[vepadulano]

Please squash commit history before merging 👍