Add openai-gpt (GPT-1) contrib model port

[dhwanw]

Description

Adds NeuronX port of OpenAI GPT-1 (117M) to the contrib models collection.

Model Information

Field	Value
Model	openai-community/openai-gpt
Architecture	Decoder-only transformer (GPT-1)
Parameters	117M
TP Degree	1
Precision	BF16

Checklist

• Model compiles successfully on Neuron
• Token matching validated (100% match, 200/200 tokens)
• Performance profiled (449.3 tok/s)
• README with architecture details, usage, validation results
• Integration tests included

Folder Structure

contrib/models/openai-gpt/
├── README.md
├── src/
│   ├── __init__.py
│   └── modeling_openai_gpt.py
└── test/
    └── integration/
        └── test_model.py

Testing

• Token Match: 100.0% (200/200 tokens, 10 prompts)
• Throughput: 449.3 tok/s (TP=1, BS=1, seq_len=128)

Compatibility

• Neuron SDK: 2.22+
• Instance: trn1.32xlarge

🤖 Generated with Claude Code