Switch eval and llama tests to use the default hip device. (#725)

Some of these workflows have been failing with
```
>           hal_device_id = haldriver.query_available_devices()[device_idx]["device_id"]
E           IndexError: list index out of range
sharktank/sharktank/utils/vmfb_runner.py:38: IndexError
```

Example logs:
https://github.com/nod-ai/shark-ai/actions/workflows/ci_eval_short.yaml?query=branch%3Amain

Rather than assume that self-hosted runners will have multiple GPUs
available and having each workflow use a specific device index, we can
use the default device and have the runners themselves choose which
devices to make visible.

.github/workflows/ci-llama-large-tests.yaml

@@ -73,7 +73,7 @@ jobs:
       - name: Run llama tests
         run: |
           source ${VENV_DIR}/bin/activate
-          pytest sharktank/tests/models/llama/benchmark_amdgpu_test.py -v -s --run-nightly-llama-tests --iree-hip-target=gfx942 --iree-device=hip://7 --html=out/llm/llama/benchmark/index.html
+          pytest sharktank/tests/models/llama/benchmark_amdgpu_test.py -v -s --run-nightly-llama-tests --iree-hip-target=gfx942 --iree-device=hip://0 --html=out/llm/llama/benchmark/index.html
 
       - name: Deploy to GitHub Pages
         uses: peaceiris/actions-gh-pages@4f9cc6602d3f66b9c108549d475ec49e8ef4d45e # v4.0.0

.github/workflows/ci_eval.yaml

@@ -70,7 +70,7 @@ jobs:
       - name: Run perplexity test with IREE
         run: |
           source ${VENV_DIR}/bin/activate
-          pytest -n 8 -v -s sharktank/tests/evaluate/perplexity_iree_test.py --run-nightly-llama-tests --bs=100 --iree-device='hip://7' --iree-hip-target=gfx942 --iree-hal-target-backends=rocm --llama3-8b-f16-model-path=/data/llama3.1/8b/llama8b_f16.irpa --llama3-8b-tokenizer-path=/data/llama3.1/8b/tokenizer_config.json --html=out/llm/llama/perplexity/iree_perplexity/index.html
+          pytest -n 8 -v -s sharktank/tests/evaluate/perplexity_iree_test.py --run-nightly-llama-tests --bs=100 --iree-device=hip://0 --iree-hip-target=gfx942 --iree-hal-target-backends=rocm --llama3-8b-f16-model-path=/data/llama3.1/8b/llama8b_f16.irpa --llama3-8b-tokenizer-path=/data/llama3.1/8b/tokenizer_config.json --html=out/llm/llama/perplexity/iree_perplexity/index.html
 
       - name: Deploy to GitHub Pages
         uses: peaceiris/actions-gh-pages@4f9cc6602d3f66b9c108549d475ec49e8ef4d45e # v4.0.0

.github/workflows/ci_eval_short.yaml

@@ -69,4 +69,4 @@ jobs:
       - name: Run perplexity test with vmfb
         run: |
           source ${VENV_DIR}/bin/activate
-          pytest -n 8 -v -s sharktank/tests/evaluate/perplexity_iree_test.py --bs=5 --iree-device='hip://6' --iree-hip-target=gfx942 --iree-hal-target-backends=rocm --llama3-8b-f16-model-path=/data/llama3.1/8b/llama8b_f16.irpa --llama3-8b-tokenizer-path=/data/llama3.1/8b/tokenizer_config.json
+          pytest -n 8 -v -s sharktank/tests/evaluate/perplexity_iree_test.py --bs=5 --iree-device=hip://0 --iree-hip-target=gfx942 --iree-hal-target-backends=rocm --llama3-8b-f16-model-path=/data/llama3.1/8b/llama8b_f16.irpa --llama3-8b-tokenizer-path=/data/llama3.1/8b/tokenizer_config.json