[Bug] Fix ndarray memory leak

[hughperkins]

Issue: #

Brief Summary

See taichi-dev/taichi#8764 by @BernardoCovas for more details.

copilot:summary

Walkthrough

copilot:walkthrough

[Copilot]

Pull Request Overview

This PR fixes a memory leak in ndarray objects by moving the __del__ method from the ScalarNdarray class to the base Ndarray class, ensuring that all ndarray types (scalar, vector, and matrix) properly clean up their underlying C++ objects when garbage collected.

Key Changes

• Moved __del__ method from ScalarNdarray to Ndarray base class to ensure all ndarray subclasses benefit from proper cleanup
• Added test case to verify that ndarray deletion properly cleans up the underlying C++ object

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 5 comments.

File	Description
python/gstaichi/lang/_ndarray.py	Moved __del__ from ScalarNdarray (removed lines 279-284) to Ndarray base class (added lines 41-45), ensuring VectorNdarray and MatrixNdarray also have proper cleanup
tests/python/test_ndarray.py	Added test_ndarray_del to verify that the C++ ndarray object is properly deleted when the Python wrapper goes out of scope

---

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

[Copilot]

Overly complex 'del' method.

Suggested change

	if impl is not None and impl.get_runtime is not None and impl.get_runtime() is not None:
	prog = impl.get_runtime()._prog
	if prog is not None:
	prog.delete_ndarray(self.arr)
	pass
	def release(self):
	if impl is not None and impl.get_runtime is not None and impl.get_runtime() is not None:
	prog = impl.get_runtime()._prog
	if prog is not None:
	prog.delete_ndarray(self.arr)