[BUG] Error with array inputs to test method

[lenabradley]

Describe the bug
When test method argument values (e.g. for parameterized inputs) are arrays, the snapshot lookup fails.

Error:

ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

Note: I believe this would also occur with other complex argument types, such as pandas dataframes.

To Reproduce

import pytest
import numpy as np

@pytest.mark.parametrize("my_array", [np.array([-1.0, 0.0, 1.0])])
def test_something(snapshot, my_array):
    """Test with array input argument"""
    snapshot.assert_match(5)

This fails on snapshot lookup, so fails on running the test for the 2nd+ time (after snapshot creation).

Error trace:

self = <snappiershot.snapshot.metadata.SnapshotMetadata object at 0x13287b250>
metadata_dict = {'arguments': {'my_array': [-1.0, 0.0, 1.0]}, 'test_runner_provided_name': '', 'update_on_next_run': False, 'user_provided_name': ''}

    def matches(self, metadata_dict: Dict) -> bool:
        """ Check if the "metadata" section of a snapshot file sufficiently matches
        this metadata object.
    
        This matching is used to identify snapshots of tests within a snapshot file.
    
        Args:
            metadata_dict: The "metadata" section to compare against this object.
        """
>       return self.caller_info.args == metadata_dict["arguments"]
E       ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

Expected behavior
No failure; correct lookup of associated snapshot

Screenshots
N/A

Versions:

• Python version:
• Snappiershot version:

Additional context
Add any other context about the problem here.

[bbonenfant]

This is an issue with the test function arguments within the metadata needing to be passed through the default_encode_value utility method. This should be relatively straightforward to fix