Graphql loaders

[wyatt-joyner-pometry]

What changes were proposed in this pull request?

Add load_nodes_from_parquet and load_edges_from_parquet GraphQL endpoints. It is specifically load_nodes_from_parquet rather than load_nodes, because I only written for this capacity at first.

Note that I am also not allowing shared_metadata, since we'll have to come up with a format for how to interpret types.

Claude mostly wrote the AppConfig changes for the allowed_parquet_paths, but there are tests for negative and positive cases of writing to allowed parquet paths.

Why are the changes needed?

We need endpoints for loading parquet files over GraphQL.

Does this PR introduce any user-facing change? If yes is this documented?

I've copied over the descriptions from the corresponding Python API.

How was this patch tested?

Added tests alongside the other existing GraphQL tests.

Are there any further changes required?

add support for shared_metadata
add support for other input formats like csv and raw blob
(more generally) add support for column name-mapping and casting

[github-actions]

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark 'Rust Benchmark'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 2.

Benchmark suite	Current: b44fbb7	Previous: 9823ef7	Ratio
lotr_graph/num_edges	4 ns/iter (± 0)	0 ns/iter (± 0)	+∞
lotr_graph/num_nodes	4 ns/iter (± 0)	1 ns/iter (± 0)	4
lotr_graph/has_node_nonexisting	5 ns/iter (± 0)	2 ns/iter (± 0)	2.50
lotr_graph/graph_latest	3 ns/iter (± 0)	0 ns/iter (± 0)	+∞
lotr_graph_materialise/materialize	8064596 ns/iter (± 26120)	1564816 ns/iter (± 35303)	5.15
lotr_graph_window_100/num_nodes	14 ns/iter (± 0)	5 ns/iter (± 0)	2.80
lotr_graph_window_100_materialise/materialize	8126806 ns/iter (± 21229)	1669150 ns/iter (± 10700)	4.87
lotr_graph_window_10/has_node_existing	142 ns/iter (± 9)	62 ns/iter (± 11)	2.29
lotr_graph_window_10_materialise/materialize	3313396 ns/iter (± 14928)	971980 ns/iter (± 4278)	3.41
lotr_graph_subgraph_10pc/has_node_nonexisting	5 ns/iter (± 0)	2 ns/iter (± 0)	2.50
lotr_graph_subgraph_10pc_materialise/materialize	1980515 ns/iter (± 14471)	334634 ns/iter (± 1287)	5.92
lotr_graph_subgraph_10pc_windowed/has_node_existing	147 ns/iter (± 10)	62 ns/iter (± 14)	2.37
lotr_graph_subgraph_10pc_windowed_materialise/materialize	1200531 ns/iter (± 16188)	230399 ns/iter (± 2617)	5.21
lotr_graph_window_50_layered/num_edges_temporal	149761 ns/iter (± 2678)	70121 ns/iter (± 7586)	2.14
lotr_graph_window_50_layered/has_node_existing	413 ns/iter (± 23)	129 ns/iter (± 12)	3.20
lotr_graph_window_50_layered/has_node_nonexisting	5 ns/iter (± 0)	2 ns/iter (± 0)	2.50
lotr_graph_window_50_layered/graph_latest	87112 ns/iter (± 1704)	36649 ns/iter (± 916)	2.38
lotr_graph_window_50_layered_materialise/materialize	30561603 ns/iter (± 527418)	3488825 ns/iter (± 24948)	8.76
lotr_graph_persistent_window_50_layered/num_edges_temporal	645419 ns/iter (± 6750)	192686 ns/iter (± 1569)	3.35
lotr_graph_persistent_window_50_layered/has_node_existing	448 ns/iter (± 403)	174 ns/iter (± 83)	2.57
lotr_graph_persistent_window_50_layered/has_node_nonexisting	5 ns/iter (± 0)	2 ns/iter (± 0)	2.50
lotr_graph_persistent_window_50_layered/iterate_exploded_edges	3476880 ns/iter (± 28692)	1659940 ns/iter (± 19402)	2.09
lotr_graph_persistent_window_50_layered/graph_latest	136183 ns/iter (± 3302)	57549 ns/iter (± 4809)	2.37
lotr_graph_persistent_window_50_layered_materialise/materialize	53423837 ns/iter (± 149257)	5298035 ns/iter (± 147912)	10.08

This comment was automatically generated by workflow using github-action-benchmark.

[fabianmurariu]

LGTM but There are a few changes needed here and there