Reject invalid Amazon Bedrock model IDs in the Anthropic provider

[guan404ming]

Why

The default model claude-opus-4-8 is a bare first-party ID that Amazon Bedrock rejects, so a Bedrock connection without an overridden model failed with an opaque 400.

Reference: https://platform.claude.com/docs/en/build-with-claude/claude-on-amazon-bedrock-legacy

How

• Add _resolve_model to validate the resolved model on the bedrock platform, raising AnthropicError with the correct prefixed-ID guidance.
• Route create_message through the validator; leave Vertex, Foundry, and first-party unchanged.
• Cover with tests for Bedrock rejection, prefixed-ID acceptance, and Vertex passthrough.

---

Was generative AI tooling used to co-author this PR?

• Yes (please specify the tool below)
  Claude Code with opus 4.8

---

• Read the Pull Request Guidelines for more information. Note: commit author/co-author name and email in commits become permanently public when merged.
• For fundamental code changes, an Airflow Improvement Proposal (AIP) is needed.
• When adding dependency, check compliance with the ASF 3rd Party License Policy.
• For significant user-facing changes create newsfragment: {pr_number}.significant.rst, in airflow-core/newsfragments. You can add this file in a follow-up commit after the PR is created so you know the PR number.

[Vamsi-klu]

Could the Bedrock check be a stricter prefix/format check instead of "anthropic." not in resolved? Any string containing that substring passes, even if it is not a valid Bedrock model ID. Checking known prefixes like global.anthropic. / region prefixes, or at least resolved.startswith(...), would better match the intended guard.

---

Drafted-by: Codex (GPT-5); reviewed by @Vamsi-klu before posting