FEAT: expm1 ufunc implementation

[SwayamInSync]

Copilot Summary

This pull request adds support for the expm1 function (which computes exp(x) - 1 with high precision) to the quad-precision and long double numerical types in the project. The implementation includes the core operation, integration with the NumPy interface, updates to documentation, and comprehensive tests to ensure correctness and precision.

Functionality additions:

• Added quad_expm1 and ld_expm1 functions to compute expm1 for quad-precision (Sleef_quad) and long double types in ops.hpp. [1] [2]
• Registered the new expm1 function with the NumPy ufunc interface, making it available for use in Python via the quad-precision type.

Documentation and testing:

• Updated the release tracker to indicate that expm1 is now implemented and tested for both quad and long double types.
• Added a comprehensive test suite for expm1 in test_quaddtype.py, covering a wide range of values including edge cases, small inputs, large inputs, and special values (inf, nan).

[juntyr]

LGMT

[SwayamInSync]

merging this in!