Added support for Multimodal eval #3996

Job	Run time
compile-gguf (macos-14)	1m 43s
gather-models-gpu	13s
gather-models-cpu	14s
runner-aoti (16-core-ubuntu)	2m 43s
runner-et (16-core-ubuntu)	4m 36s
runner-aoti (macos-14-xlarge)	2m 29s
runner-et (macos-14-xlarge)	4m 21s
test-gguf-util (macos-14)	4m 28s
test-mps (macos-m1-stable)	1m 41s
test-mps-dtype (macos-m1-stable)	1m 41s
test-torchao-aoti-experimental (macos-14-xlarge)	2m 55s
test-tinystories-executorch (16-core-ubuntu)	8m 27s
test-torchao-et-experimental (macos-14-xlarge)	9m 8s
test-tinystories-executorch (macos-14-xlarge)	8m 10s
test-torchao-experimental-mps (macos-m1-stable)	1m 59s
torchchat-command-load-test (macos-14)	1m 44s
test-build-runner-et-android / linux-job	11m 51s
test-gpu-aoti-float16 (cuda, stories15M) / linux-job	10m 35s
test-gpu-aoti-bfloat16 (cuda, stories15M) / linux-job	10m 34s
test-gpu-eval-sanity-check (cuda, stories15M) / linux-job	10m 49s
test-gpu-compile (cuda, stories15M) / linux-job	10m 20s
test-gpu-aoti-float32 (cuda, stories15M) / linux-job	10m 27s
test-cpu-eval-sanity-check (x86_64, stories15M)	4m 21s
test-cpu-aoti (x86_64, stories15M)	3m 23s
test-cpu-eval-sanity-check (aarch64, stories15M)	2m 12s
test-cpu-compile (x86_64, stories15M)	3m 33s
test-cpu-aoti (aarch64, stories15M)	2m 4s
test-cpu-compile (aarch64, stories15M)	1m 48s
test-cpu-eval-sanity-check-float16 (x86_64, stories15M)	3m 29s
test-cpu-eval-sanity-check-float32 (x86_64, stories15M)	3m 15s
test-cpu-eval-sanity-check-float16 (aarch64, stories15M)	3m 11s
test-cpu-eval-sanity-check-float32 (aarch64, stories15M)	2m 26s
	2h 30m 50s

Provide feedback