feat(v2): align client.v2 with updated ADE spec (service_tier, billing)

[yzld2002]

Summary

Aligns the (unreleased) client.v2 surface with the updated V2 AIDE Gateway OpenAPI spec. Three semantic differences were identified between the staging spec saved in #96 and the current dev spec; this PR applies all three.

Changes

1. priority → service_tier (breaking rename, but safe)
The async job-create field is renamed on parse_jobs.create and extract_jobs.create (same enum standard/priority). Done as a hard rename with no deprecation alias: client.v2 was added in #105 after the v1.12.0 tag, so priority was never in a release.

2. billing added to extract response metadata
New V2ExtractBilling { service_tier, total_credits } model, added as billing on V2ExtractMetadata (mirrors the existing V2ParseBilling). Parse metadata already modeled billing.

3. 206 / 404 on GET /v2/parse/jobs/{job_id}
No code change — the base client already treats 206 as success (envelope normalizes normally) and raises NotFoundError on 404. Added regression tests to lock the behavior.

Docs updated for the rename: api.md, examples/v2_parse.py, docs/v2-testing.md.

Out of scope

The spec also has /v2/workflow endpoints, but the SDK has no workflow resource (it never did — not one of the three diffs). Left as a follow-up.

Test plan

• TDD: updated/added tests first (red), then implemented (green)
• New tests: service_tier reaches the wire body (parse + extract), extract billing parses, parse get handles 206, parse get raises NotFoundError on 404
• ✅ Full suite: 580 passed, 240 skipped
• ✅ mypy clean on changed files; ruff clean on touched trees

🤖 Generated with Claude Code

[Copilot]

Pull request overview

This PR aligns the (unreleased) client.v2 SDK surface with updated V2 ADE gateway OpenAPI semantics by renaming the async job-create “priority” field to service_tier, modeling extract response billing metadata, and adding regression tests for parse job GET status handling.

Changes:

• Renamed async job-create parameter/body key from priority → service_tier for V2 parse and extract jobs.
• Added V2ExtractBilling and exposed it via V2ExtractMetadata.billing (and exported it from landingai_ade.types.v2).
• Added regression tests for parse job GET handling of HTTP 206 (success) and 404 (NotFoundError), plus wire-body assertions for service_tier.

Reviewed changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
tests/api_resources/v2/test_parse.py	Updates tests for service_tier rename and adds regression coverage for 206/404 on parse job GET.
tests/api_resources/v2/test_extract.py	Updates extract job create test for service_tier and adds coverage for parsing new extract metadata.billing.
src/landingai_ade/types/v2/extract_response.py	Introduces V2ExtractBilling and adds billing to V2ExtractMetadata.
src/landingai_ade/types/v2/__init__.py	Re-exports V2ExtractBilling from the public types.v2 package.
src/landingai_ade/resources/v2/parse.py	Renames parse job-create parameter/body field to service_tier.
src/landingai_ade/resources/v2/extract.py	Renames extract job-create parameter/body field to service_tier.
examples/v2_parse.py	Updates example usage to service_tier.
api.md	Updates API docs for service_tier rename and documents extract billing type nesting.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.