Optimize TCQ with block-level AVX2 dispatch

[chengchen-google]

Previously, TCQ evaluated coefficients along anti-diagonal scan lines in a portable C loop (trellis_loop_diagonal_st8), invoking individual fine-grained RTCD SIMD functions per coefficient.

This fine-grained dispatch structure incurred significant overhead:

• Repeated RTCD function calls and indirect branches on every coeff.
• Register spills to memory structures across subroutine boundaries.
• Optimization barriers preventing compiler vector scheduling across the full diagonal loop.

This patch introduces a coarse-grained, block-level RTCD entry point:
av2_trellis_loop_diagonal_st8()
with AVX2 specialization (av2_trellis_loop_diagonal_st8_avx2).

Updates are now executed contiguously within AVX2 registers without exiting to scalar C code per coefficient.

Unit tests are added to compare the the new and original implementation.

We can see that the TCQ function becomes 3-10% faster.

=======================================================================
TX Size   Pure C (us)   Base AVX2 (us)   Patch AVX2 (us)  vs Base AVX2
=======================================================================
4x4       110313        45646            44360           1.03x (+2.8%)
8x8       369747        102636           96537           1.06x (+5.9%)
16x16     343618        81354            74201           1.10x (+8.8%)
32x32     268794        60375            54646           1.10x (+9.5%)
4x8       195707        64939            61781           1.05x (+4.9%)
8x4       201683        65227            61872           1.05x (+5.1%)
8x16      175484        44150            40669           1.09x (+7.9%)
16x8      180465        44455            40943           1.09x (+7.9%)
16x32     134924        30975            28011           1.11x (+9.6%)
32x16     134311        30946            28192           1.10x (+8.9%)
4x16      364581        103343           96494           1.07x (+6.6%)
16x4      390933        104326           97702           1.07x (+6.3%)
8x32      343005        81586            74379           1.10x (+8.8%)
32x8      352547        81916            74820           1.09x (+8.7%)
=======================================================================

[jjustiss-apple]

LGTM, just a minor unit test maintenance question.

[urvangjoshi]

@chengchen-google : This is the unit test that is failing:

[ RUN      ] AVX2/TcqLoopDiagonalSt8Test.RandomValues/0
../../test/trellis_test.cc:1117: Failure
Expected equality of these values:
  rate_ref
    Which is: -1099013650
  rate_tst
    Which is: 193634

[  FAILED  ] AVX2/TcqLoopDiagonalSt8Test.RandomValues/0, where GetParam() = bit_depth:0 function:0x59d01f66a390 function:0x59d01fafb590 (0 ms)
[----------] 1 test from AVX2/TcqLoopDiagonalSt8Test (0 ms total)

For reference, this section of wiki page explains how to find the failing test: https://github.com/AOMediaCodec/avm/wiki/Reproducing-CI-Test-Failures-Locally#finding-failing-tests