4x4 LD64 GEMM microkernel in AArch32+VFP assembly

PiperOrigin-RevId: 312008451
diff --git a/BUILD.bazel b/BUILD.bazel
index 275ff2a..7ae306b 100644
--- a/BUILD.bazel
+++ b/BUILD.bazel
@@ -1561,6 +1561,7 @@
 ]
 
 AARCH32_ASM_UKERNELS = [
+    "src/f32-gemm/4x4-aarch32-vfp-ld64.S",
     "src/f32-gemm/4x4-minmax-aarch32-vfp-ld64.S",
     "src/f32-gemm/4x8-minmax-aarch32-neon-cortex-a53.S",
     "src/f32-gemm/4x8-minmax-aarch32-neon-cortex-a55.S",