chore(terraform): Update models

[PabloZaiden]

This pull request updates the available AI model outputs and deployment defaults across the common_models module and all reference architectures. The main focus is on replacing older GPT-4.x and O4 models with the latest GPT-5 series and new audio/transcription models, while also expanding the documentation to reflect these new capabilities.

Model output and deployment updates:

• Replaced all references to gpt_4_1 and o4_mini with newer GPT-5 models (such as gpt_5, gpt_5_2_chat, and gpt_5_nano) in module defaults and reference architectures, ensuring deployments use the latest models. [1] [2] [3] [4] [5] [6]
• Updated the SKU override example in documentation to use gpt_5 instead of gpt_4_1.

Expanded and improved model outputs:

• Added outputs for new GPT-5 variants (gpt_5, gpt_5_mini, gpt_5_nano, gpt_5_chat, gpt_5_1, gpt_5_1_chat, gpt_5_1_codex, gpt_5_1_codex_mini, gpt_5_2, gpt_5_2_chat) and new audio models (gpt_audio, gpt_audio_mini, gpt_realtime, gpt_realtime_mini, gpt_4o_mini_transcribe, gpt_4o_transcribe_diarize). [1] [2]
• Updated descriptions for all outputs to clearly state their purpose and capabilities. [1] [2]

Documentation improvements:

• Overhauled the outputs table in README.md to document all new models and provide more detailed descriptions, including new transcription and diarization models.

These changes collectively ensure that all modules and reference architectures are aligned with the latest AI model offerings and that documentation is clear and up to date.

[Copilot]

Pull request overview

This PR modernizes the AI model catalog by replacing older GPT-4.x and O4 models with the latest GPT-5 series models, introducing new audio and transcription capabilities, and updating all reference architectures to use these new defaults.

Key Changes:

• Replaced GPT-4.1 and O4-mini with GPT-5.2-chat and GPT-5-nano as default deployments across all four reference architectures
• Added 10 new GPT-5 model variants (including chat, codex, mini, and nano options) and 6 new audio/transcription models to the common_models catalog
• Enhanced model output descriptions to provide clear capability summaries and use case guidance

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
modules/common_models/outputs.tf	Replaces all GPT-4.x/O4 model outputs with comprehensive GPT-5 series and adds new audio/transcription model outputs
modules/common_models/README.md	Updates documentation table to reflect new model catalog and updates usage examples to reference GPT-5
modules/ai_foundry/README.md	Updates example to use gpt_5 instead of gpt_4_1
reference_architectures/foundry_basic/main.tf	Updates default model deployments to use gpt_5_2_chat and gpt_5_nano
reference_architectures/foundry_basic_private/main.tf	Updates default model deployments to use gpt_5_2_chat and gpt_5_nano
reference_architectures/foundry_standard/main.tf	Updates default model deployments to use gpt_5_2_chat and gpt_5_nano
reference_architectures/foundry_standard_private/main.tf	Updates default model deployments to use gpt_5_2_chat and gpt_5_nano