AnthropicChatModel exposes `internalStream` method publicly

While doing some API mapping of the Spring AI project for my job (I work in security research @ Veracode) I noticed that the Anthropic implementation of ChatModel appears to expose a method that has a naming prefix that would indicate it probably is not intended to have public access:

https://github.com/spring-projects/spring-ai/blob/9b76ba9efca3fd07633d3d90b6163a79f0c05acc/models/spring-ai-anthropic/src/main/java/org/springframework/ai/anthropic/AnthropicChatModel.java#L307

Checking other implementations, and just perusing the code a bit It seems that this was mentioned in an update guide:

https://github.com/spring-projects/spring-ai/blob/9b76ba9efca3fd07633d3d90b6163a79f0c05acc/spring-ai-docs/src/main/antora/modules/ROOT/pages/upgrade-notes.adoc#the-access-level-of-model-internal-methods-changed-to-private

I was going to open a PR, but the guide says to avoid opening a PR for straightforward changes, and this looks like it meets that criteria.

Cheers!

Open models/spring-ai-anthropic/src/main/java/org/springframework/ai/anthropic/AnthropicChatModel.java at the referenced internalStream method and compare it with the access-level guidance in spring-ai-docs/src/main/antora/modules/ROOT/pages/upgrade-notes.adoc. Done means AnthropicChatModel no longer exposes internalStream publicly.