Implement tabular benchmarks in Python and C++

[ashvardanian]

The benchmark will be receiving a directory of Parquet, CSV, Avro or Orca files and exporting them simultaneously into graph, document, and, potentially, path collections. The latter can also be added into the twitter benchmark.

[ashvardanian]

Efficient Python implementation would require finishing #41

[ashvardanian]

The construction part may be just a direct call to ukv_docs_import and ukv_graph_import_t from #154.
After the construction, sampling (#153) should be performed to randomly selected items to be queried in the analytical part of the benchmarks. Over time that benchmark will replace the twitter benchmark.