#6344: Update RoBERTa QA demo

[kkeerthana0573]

Ticket

• Link to Github Issue

Problem description

• Functional RoBERTa model Demo.
• Port RoBERTa model to n300.

What's changed

• TTNN support for RoBERTa model is done using ttnn_optimized_bert pipeline.
• Ported RoBERTa model functionality to n300.
• Added demo for RoBERTaForQuestionAnswering task.

Checklist

• Post commit CI passes
• Blackhole Post commit (if applicable)
• Model regression CI testing passes (if applicable)
• Device performance regression CI testing passes (if applicable)
• New/Existing tests provide coverage for changes

[eyonland]

Looking for tests with both these annotations...

@pytest.mark.models_device_performance_bare_metal
@pytest.mark.models_performance_bare_metal

Note that we document this here: https://tenstorrent.github.io/tt-metal/latest/ttnn/ttnn/demos.html
See for example:
tests/ttnn/integration_tests/resnet/test_performance.py

[kkeerthana0573]

@eyonland,
Will add the test and update the PR.
Thank you.

[tt-rkim]

This should be good.

Please re-run the GS branch of ttnn integration tests in FD nightly.

I see

FAILED tests/ttnn/integration_tests/roberta/test_ttnn_optimized_roberta.py::test_roberta[sequence_size=384-batch_size=8-model_name=deepset/roberta-large-squad2] - OSError: We couldn't connect to 'https://huggingface.co' to load this file, couldn't find it in the cached files and it looks like deepset/roberta-large-squad2 is not the path to a directory containing a file named config.json.

WH passes, but want to make sure of GS.

[kkeerthana0573]

@tt-rkim,
FD Nightly for GS passed. Please find the link here.

[tt-rkim]

Great, thank you

[saichandax]

@esmalTT , Could you please review and give your approval for this PR?

[esmalTT]

Do these PCC bounds need to be updated?