feat: add dynamic model registration support to TGI inference

[mattf]

What does this PR do?

adds dynamic model support to TGI

add new overwrite_completion_id feature to OpenAIMixin to deal with TGI always returning id=""

Test Plan

tgi: docker run --gpus all --shm-size 1g -p 8080:80 -v /data:/data ghcr.io/huggingface/text-generation-inference --model-id Qwen/Qwen3-0.6B

stack: TGI_URL=http://localhost:8080 uv run llama stack build --image-type venv --distro ci-tests --run

test: ./scripts/integration-tests.sh --stack-config http://localhost:8321 --setup tgi --subdirs inference --pattern openai

[Copilot]

Pull Request Overview

This PR adds dynamic model registration support to Text Generation Inference (TGI) provider by implementing OpenAI-compatible endpoints and introducing a new feature to handle TGI's empty response IDs.

Key changes:

• Enhanced TGI provider with OpenAI compatibility for dynamic model registration
• Added overwrite_completion_id feature to handle providers that return empty IDs
• Integrated TGI support into test infrastructure with new test setup and recording files

Reviewed Changes

Copilot reviewed 12 out of 14 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
tests/integration/suites.py	Added "tgi" test setup configuration for integration testing
tests/integration/recordings/responses/*.json	Test recording files capturing TGI API responses for various scenarios
tests/integration/inference/test_openai_completion.py	Removed TGI from exclusion lists, allowing OpenAI completion tests
llama_stack/providers/utils/inference/openai_mixin.py	Added overwrite_completion_id feature and ID generation logic
llama_stack/providers/remote/inference/tgi/tgi.py	Refactored to inherit from OpenAIMixin and support dynamic model registration

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}