feat: add T-Systems LLMHub as a provider

[ashwinnair14]

Summary

• Adds T-Systems LLMHub (AI Foundation Services) as a new provider with 38 models
• Uses @ai-sdk/openai-compatible with API endpoint https://llm-server.llmhub.t-systems.net/v2
• Includes LLMs (Claude, GPT, Gemini, Llama, Mistral, Qwen, o-series), embeddings, audio (Whisper), and image generation models
• T-Systems offers GDPR-compliant, EU-hosted access to both open-source and proprietary models

Models included

Category	Count	Examples
LLMs	26	GPT-4.1, GPT-5, Claude Sonnet 4/4.5, Gemini 2.5/3, o3/o4-mini, Llama 3.3, Mistral, Qwen3
Embeddings	5	Jina v2, BGE M3, Ada 002, TSI ColQwen2
Audio	2	Whisper Large V3, Whisper Large V3 Turbo
Image	1	GPT Image 1
Specialty	1	Teuken 7B (EU multilingual)
Regional	1	GPT-OSS 120B (T-Cloud Germany)

Pricing

Pricing sourced from T-Systems documentation (Standard1000 plan tier). Open-source models hosted on T-Cloud Germany use flat-rate pricing; proprietary models have per-token input/output rates.

Data sources

• Provider docs: https://docs.llmhub.t-systems.net/
• Model IDs and pricing: telekom/aifoundation-docs pricing config v1.20.2
• Model specs cross-referenced from existing provider entries in this repo

Test plan

• bun validate passes with exit code 0
• All 38 model TOML files created with required schema fields
• Provider TOML includes required api field for @ai-sdk/openai-compatible
• Logo SVG uses currentColor with no fixed sizes

[ashwinnair14]

Draft stage.