chore(ty): replace pyright with ty in RBMDS

[sahiljhawar]

No description provided.

[Copilot]

Pull request overview

This PR updates RBMDataSet-related Python code and project configuration to use the ty type checker instead of Pyright, primarily by adjusting ty configuration and adding ty:ignore[...] suppressions where ty currently lacks inference/stub coverage.

Changes:

• Updated pyproject.toml to include swvo/io/RBMDataSet in ty checking (removed it from exclude).
• Replaced/added ty:ignore[...] directives across RBMDataSet modules and scripts to address new ty diagnostics.
• Cleaned up some Pyright-specific ignores and tightened some type annotations (e.g., Literal["mat", "pickle"]).

Reviewed changes

Copilot reviewed 9 out of 9 changed files in this pull request and generated 9 comments.

Show a summary per file

File	Description
swvo/io/RBMDataSet/utils.py	Adds ty suppressions around pandas datetime conversion and timezone handling.
swvo/io/RBMDataSet/scripts/create_RBSP_line_data.py	Adds ty suppressions in line-data creation logic and RBMDataSet attribute usage.
swvo/io/RBMDataSet/interp_functions.py	Adds Literal import and ty suppressions in interpolation functions.
swvo/io/RBMDataSet/identify_orbits.py	Removes Pyright-specific ignore comments for SciPy usage.
swvo/io/RBMDataSet/bin_and_interpolate_to_model_grid.py	Adds ty suppressions for typing issues in binning, plotting, and multiprocessing progress reporting.
swvo/io/RBMDataSet/RBMNcDataSet.py	Replaces broad type: ignore with ty-specific ignores for assignments and argument types.
swvo/io/RBMDataSet/RBMDataSetManager.py	Tightens preferred_extension to a Literal and adds a ty suppression for satellite typing.
swvo/io/RBMDataSet/RBMDataSet.py	Adds ty suppressions, removes some type: ignore, and tweaks inline ignore/formatting.
pyproject.toml	Enables ty checking for RBMDataSet by removing it from the excluded paths.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

loaded_var_arrs is annotated as dict[str, NDArray[np.number]], but joined_value may not always match that type, leading to ty:ignore[invalid-assignment]. Rather than suppressing, adjust the dict’s value type (e.g., include NDArray[np.object_] / NDArray[np.generic]) or refactor to keep numeric vs non-numeric entries separate.

[Copilot]

Using assert for argument validation can be skipped when Python is run with optimizations (-O), which would remove this check. Prefer an explicit if ...: raise ValueError(...) here for user-facing input validation. Updating the types after normalization (e.g., cast target_en/target_al to list[float]) should also remove the need for the ty:ignore on this line.

[Copilot]

preferred_extension is now restricted to Literal["mat", "pickle"], but the overloads above still accept str, so type checkers will allow unsupported values. Either update the overload signatures to the same Literal[...] (and consider validating/raising on unexpected values) or revert the annotation to str with runtime validation.

[Copilot]

The ty:ignore directives on these lines are duplicated and include ignore-comment-unknown-rule, which is not a valid diagnostic code. This can cause ty to reject the ignore comment and fail the type-check. Consolidate to a single ty:ignore[...] with only valid codes (or fix the underlying type issue) and remove the unknown rule name.

Suggested change

	grid_X = grid_R[:, :, 0, 0] * np.cos(grid_P[:, :, 0, 0]) # ty:ignore[non-subscriptable] # ty:ignore[ignore-comment-unknown-rule, not-subscriptable]
	grid_Y = grid_R[:, :, 0, 0] * np.sin(grid_P[:, :, 0, 0]) # ty:ignore[non-subscriptable] # ty:ignore[ignore-comment-unknown-rule, not-subscriptable]
	grid_X = grid_R[:, :, 0, 0] * np.cos(grid_P[:, :, 0, 0]) # ty:ignore[non-subscriptable]
	grid_Y = grid_R[:, :, 0, 0] * np.sin(grid_P[:, :, 0, 0]) # ty:ignore[non-subscriptable]