Fix CI failure on Python 3.13/3.14 with numpy 2.4.0+ (#929)

Add concise __repr__ methods to ErrorSnapshot and PropagatedErrorSnap…

…shot (#927)

The default dataclass __repr__ shows all fields, making output very verbose
when these objects appear in arrays or other containers. Add custom __repr__
methods that show only essential information:

- ErrorSnapshot: function name, exception type and message
- PropagatedErrorSnapshot: function name and reason

The detailed __str__ output remains unchanged for when users explicitly
print() these objects.

ENH: `error_handling` options (`raise` and `continue`) (#854)

⬆️ Update actions/checkout action to v6 (#923)

Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>

DOC: Add missing 'Function Chaining Helper' in index (#918)

DOC: Add comparison section with Hamilton (#502)

BUG: Fix path resolution and add legacy format support for RunInfo.lo…

…ad (#898)

Changes `RunInfo` to store paths relative to `run_folder` for portability. Adds `_legacy_fix()` to handle data from versions <=0.86.0 where paths included the `run_folder` prefix.

When running e.g., `pipeline.map(inputs, run_folder="foo/my_run_folder")`:

New format (>v0.86.0):
- `run_folder: /absolute/path/to/run_folder`
- `input_paths: {"x": "inputs/x.cloudpickle"}`
- `defaults_path: "defaults/defaults.cloudpickle"`

Legacy format (<=v0.86.0):
- `run_folder: "foo/my_run_folder"`
- `input_paths: {"x": "foo/my_run_folder/inputs/x.cloudpickle"}`
- `defaults_path: "foo/my_run_folder/defaults/defaults.cloudpickle"`

This enables loading `run_folders` from different working directories while maintaining backward compatibility with existing data.

ENH: Extract type annotations from NamedTuple return types (#874)

When a function returns a NamedTuple and has multiple output names,
PipeFunc now correctly extracts the individual field type annotations
from the NamedTuple instead of returning NoAnnotation.

This allows for proper type checking and better IDE support when using
NamedTuples as return types with multiple outputs.

ENH: More comprehensive default value comparison with `is_equal` (#860)

* Using is_equal function in default values comparison

* Fix test names

* Raise when comparing defaults

---------

Co-authored-by: Bas Nijholt <[email protected]>

BUG: Cast to `bool` to prevent issues with custom equality methods (#859

)