feat(ai): add LiteLLM adapter

[nam157]

Summary

• add a LiteLLM AI provider adapter using the OpenAI-compatible chat completions API
• support configurable LiteLLM proxy base URLs while defaulting to the local proxy
• cover dry-run behavior, request mapping, usage mapping, and error excerpts with Vitest

Closes #266
Related to #133

Verification

• pnpm exec vitest run packages/ai/litellm/src/index.test.ts
• pnpm --filter @profullstack/sh1pt-ai-litellm typecheck
• pnpm --filter @profullstack/sh1pt-ai-litellm build
• pnpm install --frozen-lockfile
• git diff --check

[github-actions]

🤖 Auto-rebase: The branch was rebased successfully locally but could not be pushed to the fork. Please enable 'Allow edits from maintainers' in the PR settings, or rebase manually: git fetch upstream master && git rebase upstream/master.