docs(a2a): lower business agent Python prerequisite to 3.10

[FanouZeng-TT]

Description

a2a/business_agent/README.md lists Python 3.13 as a prerequisite, but the
sample's pyproject.toml declares requires-python = ">=3.10", and running
the CLI under CPython 3.10 works:

uv run --python 3.10 business_agent --help
# exit 0, prints the CLI usage

The 3.13 requirement is a development convenience recorded in the sample's
.python-version file, not a compatibility promise for users; the README
drives the version new users install, so it should match the package metadata.

Fix: lower the documented prerequisite from Python 3.13 to Python 3.10+
to match requires-python = ">=3.10".

Category (Required)

• Core Protocol: Protocol specification changes. (Requires Technical Council approval)
• Governance/Contributing: Governance or contribution process changes. (Requires Governance Council approval)
• Capability: Capability specification changes. (Requires Maintainer approval)
• Documentation: Documentation updates. (Requires Maintainer approval)
• Infrastructure: Build, CI, tooling, or release changes. (Requires DevOps Maintainer approval)
• Maintenance: Dependency, compatibility, or routine maintenance changes. (Requires DevOps Maintainer approval)
• SDK: Language-specific SDK updates and releases. (Requires DevOps Maintainer approval)
• Samples / Conformance: Maintaining samples and the conformance suite. (Requires Maintainer approval)
• UCP Schema: Changes to the ucp-schema tool. (Requires Maintainer approval)
• Community Health (.github): Repository-wide health files and configuration. (Requires DevOps Maintainer approval)

Related Issues

N/A

Checklist

• I have followed the Contributing Guide.
• I have updated the documentation (if applicable).
• My changes pass all local linting and formatting checks.
• I have added tests that prove my fix is effective or that my feature works.
• New and existing unit tests pass locally with my changes.
• (For Core/Capability) I have included/updated the relevant JSON schemas.
• I have regenerated Python Pydantic models by running generate_models.sh under python_sdk.

Screenshots / Logs (if applicable)

uv run --python 3.10 business_agent --help: exit 0, usage output
uvx pre-commit run --all-files: passed
git diff --check: passed