providers: consolidate openai providers

[saem]

There are two providers one normal and one embedding:

1. OpenAIProvider
2. OpenAIEmbeddingProvider

These can be combined with a little bit of logic, that would reduce redundant code paths.

This was identified as a good future refactoring during PR: #185

[rohitg00]

Agreed — this would simplify both providers and reduce the surface area for embedding-vs-chat config divergence. Happy to take a PR.

Suggested shape (open to alternatives):

src/providers/openai.ts exports a single OpenAIProvider class with two entry points:

• compress(text, opts) → chat-completions (current OpenAIProvider)
• embed(texts, opts) → embeddings endpoint (current OpenAIEmbeddingProvider)

Shared: base URL resolution (OPENAI_BASE_URL), API-key handling, retry/circuit-breaker wiring, error normalization. Both endpoints are POST-only with the same auth header, so the request infra collapses cleanly.

One thing to watch: OPENAI_EMBEDDING_MODEL and the chat OPENAI_MODEL should stay independent env vars — collapsing them was a real footgun I want to avoid.

Tagging as good-first-issue for whoever picks this up. Reference: PR #185 introduced the second class, so the diff there is the right starting point for what to merge back together.