Graphistry

Fixes

* `Plottable._render` now typed as `RenderModesConcrete`
* Remote GFQL - Handle `output_type is None`

Docs

* Rename CONTRIBUTE.md to CONTRIBUTING.md to match OSS standards (snyk)
* setup.py: add project_urls
* Add FUNDING.md
* Add CODE_OF_CONDUCT.md

Feat

* GFQL serialization: Edge AST node deserializes as more precise `ASTEdge` subclasses

Fixes

* GFQL Hop: Detect 614 of node id column name colliding with edge src/dst id column name and raise `NotImplementedError`
* MyPy: Remove explicit type annotation from Engine

Docs

* Python remote mode notebook: Fixed engine results
* Python remote mode: Add JSON example

Fixes

* Fix broken imports in new GFQL remote mode

Docs

* New tutorial on GPU memory management and capacity planning in the GPU section
* New tutorial on remote-mode GFQL
* New tutorial on remote-mode Python

Feat

* `plot(render=)` supports literal-typed mode values: `"auto"`, `"g"`, `"url"`, `"ipython"`, `"databricks"`, where `"g"` is a new Plottable
* Remote metadata: Expose and track uploaded `._dataset_id`, `._url`, `._nodes_file_id`, and `._edges_file_id`
* Remote upload: Factor out explicit upload method `g2 = g1.upload(); assert all([g2._dataset_id, g2._nodes_file_id, g2._edges_file_id])` from plot interface
* Remote dataset: Remote dataset binding via `g1 = graphistry.bind(dataset_id='abc123')`
* Remote GFQL: Remote GFQL calls via `g2 = g1.chain_remote([...])` and `meta_df = g1.chain_remote_shape([...])`
* Remote GPU Python: Remote Python calls via `g2 = g1.python_remote_g(...)`, `python_remote_table()`, and `python_remote_json()` for different return types

Changed

* `plot(render=)` now `Union[bool, RenderMode]`, not just `bool`

Added

* Layout: `circle_layout()` that moves points to one or more rings based on an ordinal field such as degree
* Layout: `fa2_layout()` that uses the ForceAtlas2 algorithm to lay out the graph, with some extensions.
* Layout: `group_in_a_box_layout()` is significantly faster in CPU and GPU mode
* Layout: `group_in_a_box_layout()` exposes attribute `_partition_offsets`
* Compute methods `g.to_pandas()` and `g.to_cudf()`

Fix

* cugraph methods now handle numerically-typed node/edge index columns
* repeat calls to `fa2_layout` and `group_in_a_box_layout` now work as expected

Docs

* Add group-in-a-box layout tutorial

Infra

* Expose `scene_settings` in `Plottable`