[provider-anthropic] Support 1h prompt-cache TTL — verified working against the live API, no beta header required

Summary

provider-anthropic hardcodes cache_control: {"type": "ephemeral"} (5-minute TTL) at all three breakpoint sites (__init__.py:2136, :3719, :3745, verified on main 2026-07-20). There is no config to request the 1-hour TTL.

For human-paced interactive sessions, any gap >5 minutes between turns expires the cache and the next call re-writes the FULL context at cache-write rates. Observed: a single 282k-token cache-write event after one thinking pause (session data in the companion loop-streaming issue). With large system prompts (bundles + MCP tool schemas commonly reach 130k+ tokens base), these expiry re-writes dominate cost.

Verified against the live API (2026-07-20)

ttl: "1h" is accepted today with no beta header:

• cache_control: {type: ephemeral, ttl: "1h"} -> HTTP 200
• usage returns a per-TTL breakdown: cache_creation: {ephemeral_5m_input_tokens: 0, ephemeral_1h_input_tokens: 4419}

This also means the comment in _cost.py:131 ("Anthropic's usage object returns a single cache_creation_input_tokens count") is stale — the API now distinguishes tiers, so 1h writes (2x base input) can be cost-tracked correctly.

Suggested change

1. New provider config key cache_ttl: "5m" | "1h" (default "5m", preserving current behavior), applied at all three breakpoint sites.
2. _cost.py: read the cache_creation per-TTL breakdown and price 1h writes at 2x base input.

Economics: 1h writes cost 2x vs 1.25x, but for sessions with several >5min gaps per hour, converting full-context re-writes (1.25x each) into cache reads (0.1x) is a large net win.

Happy to submit the PR — small, config-gated, default-off.

Related

• Companion issue: https://github.com/microsoft/amplifier/issues/336
• Existing PRs #58 / #65 on amplifier-module-provider-anthropic cover missing Fable rates in _RATES; the per-TTL cost handling here composes with those.

Start with the three cache_control breakpoint sites in init.py at lines 2136, 3719, and 3745, then read _cost.py:131 and the provider configuration handling. Done means a default 5m setting preserves current behavior, 1h requests are applied at every breakpoint, and the per-TTL cache_creation usage is priced correctly.