feat: multi-provider routing + /v1/usage reporting endpoint

[pyramation]

Summary

Adds multi-provider routing + /v1/usage reporting endpoint to the agentic server, and extracts the MetaSchema-based table resolution into a shared @constructive-io/usage-loader workspace package.

Multi-provider routing:

• Routes by X-Provider header, model prefix (anthropic/claude-3.5-sonnet), or known patterns (claude-* → anthropic, gpt-* → openai, llama-* → ollama)
• Anthropic fully supported: x-api-key auth, system message extraction, response transforms to OpenAI format
• GET /v1/providers — lists configured providers
• POST /v1/usage — fire-and-forget endpoint for Python functions to report inference usage

@constructive-io/usage-loader package:

// Before: hardcoded schema in 3 separate files
pool.query(`INSERT INTO "constructive_usage_public".platform_usage_log_computes ...`)

// After: single canonical loader, MetaSchema-resolved, 60s TTL cache
import { UsageLoader } from '@constructive-io/usage-loader';
const loader = new UsageLoader(pool, databaseId);
loader.logComputeUsage(entry);   // resolves table from metaschema_public.table
loader.logInferenceUsage(entry);
loader.logStorageUsage(entry);

All three consumers (worker/compute-meter, agentic-server/inference-meter, fn-runtime/storage) now import from this shared package. job/worker/usage-client.ts is a thin re-export for backward compat.

Tests: 119 integration + 25 unit, all green.

Link to Devin session: https://app.devin.ai/sessions/b2291a8e333e445aa125a2efd1996206
Requested by: @pyramation

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment, CI, and merge conflict monitoring