llms/anthropic: handle multiple parts in AI messages

[Rony-ZenAlden]

handleAIMessage only inspected msg.Parts[0], so an assistant message containing both text and a tool call (e.g. [TextContent, ToolCall]) serialized only the first part and silently dropped the rest. This caused Anthropic API errors on the following turn because the tool_result had no corresponding tool_use block.

Iterate over every part and emit a content block for each, mirroring handleHumanMessage. This also fixes a potential panic when Parts is empty (previously an out-of-range access on Parts[0]).

Add unit tests covering a mixed text + tool-call message and the empty-parts case.

Fixes #1468

PR Checklist

• Read the Contributing documentation.
• Read the Code of conduct documentation.
• Name your Pull Request title clearly, concisely, and prefixed with the name of the primarily affected package you changed according to Good commit messages (such as memory: add interfaces for X, Y or util: add whizzbang helpers).
• Check that there isn't already a PR that solves the problem the same way to avoid creating a duplicate.
• Provide a description in this PR that addresses what the PR is solving, or reference the issue that it solves (e.g. Fixes #123).
• Describes the source of new concepts.
• References existing implementations as appropriate.
• Contains test coverage for new functions.
• Passes all golangci-lint checks.