genai: define cached tokens attributes

[verdie-g]

Fixes #1959

Changes

This PR adds two new GenAI attributes to represent provider-level prompt caching:

• gen_ai.usage.cache_read_input_tokens
• gen_ai.usage.cache_creation_input_tokens

It also updates the description of gen_ai.usage.input_tokens to state that it must include cached tokens. OpenAI and Vertex AI already count cached tokens in input_tokens, while Anthropic excludes them. Without clarification, instrumentation would produce incompatible values. Requiring input_tokens to always include cached tokens ensures a consistent, cross-provider definition of total input tokens, with the new attributes exposing the cached breakdown.

Merge requirement checklist

• CONTRIBUTING.md guidelines followed.
• Change log entry added, according to the guidelines in When to add a changelog entry.
  • If your PR does not need a change log, start the PR title with [chore]
• Links to the prototypes or existing instrumentations (when adding or changing conventions)

[alexmojaki]

I think opt_in refers to the user opting in to instrumenting this attribute, which shouldn't be necessary. I don't think it's about whether the user has enabled the relevant feature.