Integrate lossless forward transform intrinsics #4707

Job	Run time
rustfmt-clippy	2m 28s
build-macos (Cargo build (x64), cargo-build, x86_64-apple-darwin)	4m 53s
build-unix (beta-build)	3m 27s
build-macos (Cargo test (x64), cargo-test, x86_64-apple-darwin)	4m 37s
build-unix (1.70.0-tests)	6m 10s
build-macos (Cargo C-build (x64), cargo-c, x86_64-apple-darwin)	10m 35s
build-unix (aom-tests)	8m 52s
build-macos (Cargo build (Arm64), cargo-build, aarch64-apple-darwin)	4m 5s
build-windows (cargo-build, x86_64-pc-windows-msvc)	7m 49s
build-unix (dav1d-tests)	7m 49s
build-macos (Cargo C-build (Arm64), cargo-c, aarch64-apple-darwin)	6m 7s
build-windows (cargo-test, x86_64-pc-windows-msvc)	11m 49s
build-unix (no-asm-tests)	10m 33s
build-unix (grcov-codecov)	12m 48s
build-windows (cargo-c, x86_64-pc-windows-gnu)	11m 13s
build-unix (bench)	4m 50s
build-unix (doc)	3m 10s
build-unix (cargo-c)	7m 14s
build-unix (check-no-default)	1m 43s
build-unix (check-extra-feats)	2m 47s
build-unix (check-unstable-feats)	2m 43s
build-unix (fuzz)	5m 9s
build-unix (dav1d-tests-arm64, stable)	12m 17s
	2h 33m 8s

Provide feedback