VX4B remaining test failures fix

[andrewstanfordjason]

No description provided.

[Copilot]

Pull request overview

This PR fixes remaining test failures for the VX4B architecture by addressing a critical test bug, adjusting test tolerances for VX4B precision differences, and fixing assembly instruction scheduling issues.

Changes:

• Fixed array indexing bug in test_vect_mul.c where A[0] was checked instead of A[i] in a loop
• Added VX4B-specific test tolerance assertions for s16 multiplication and complex s32-to-s16 conversion tests
• Separated potentially conflicting assembly instructions in dct_adsb_s32.S to avoid pipeline hazards
• Clarified stack size calculation in chunk_s32_log.S (functionally equivalent: 8+8*6 = 8+48)

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated no comments.

File	Description
tests/vect_tests/src/vect/test_vect_mul.c	Fixes critical bug where A[0] was tested instead of A[i] in loop; adds VX4B tolerance (±4) for s16 mul tests
tests/vect_tests/src/vect/complex/test_vect_complex_s32_to_complex_s16.c	Adds VX4B tolerance (±1) for complex s32-to-s16 conversion tests
lib_xcore_math/src/arch/vx4b/dct/s32/dct_adsb_s32.S	Separates instruction bundle to avoid potential pipeline hazard with t3 register
lib_xcore_math/src/arch/vx4b/chunk_s32/chunk_s32_log.S	Makes stack size calculation more explicit (8+8*6 vs 8+48, same value)

---

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