fix(confluence): populate metadata for page children

[Cchena1]

Description

confluence_get_page_children currently requests only version by default, so
the Confluence API response does not include the history fields used to populate
created, updated, and author.

Request history alongside version by default and keep version.when as the
fallback update timestamp when history is unavailable.

This follows the metadata expansion pattern previously addressed in #1117 and
#1480.

Fixes: #1602

Changes

• Expand version,history by default when retrieving child pages.
• Use version.when as the update timestamp fallback even when history is absent.
• Add regression coverage for child-page metadata and version-only responses.
• Update the generated Confluence tool documentation.

Testing

• Unit tests added/updated
• Integration tests passed
• Manual checks performed: live Data Center response populated created, updated, and author; the shared Cloud hierarchy path also passed E2E.

Commands run:

• uv run pytest tests/unit/models/test_confluence_page_model.py tests/unit/confluence/test_pages.py tests/unit/servers/test_confluence_server.py -xq
  • 226 passed
• uv run pytest tests/integration/test_real_api.py::TestRealConfluenceAPI::test_page_hierarchy --integration --use-real-data -xv -ra
  • 1 passed
• uv run pytest tests/e2e/test_confluence_dc_operations.py::TestConfluenceDCPageHierarchy::test_create_child_page --dc-e2e -xv -ra
  • 1 passed
• uv run pytest tests/e2e/cloud/test_confluence_cloud_operations.py::TestConfluenceCloudPageHierarchy::test_create_child_page --cloud-e2e -xv -ra
  • 1 passed
• Live Data Center metadata assertion
  • created, updated, and author populated
• uv run python scripts/generate_tool_docs.py --check
• Ruff format/lint and mypy checks on all changed Python files
• git diff --check origin/main...HEAD

Maintainer verification (2026-08-20)

• Ran the full unit suite: 3839 passed; the five skips are opt-in real-data model tests, and the targeted regression suite ran without skips (226 passed).
• Re-ran the real Confluence hierarchy integration test with Cloud credentials: 1 passed with no skips.
• Re-ran Data Center and Cloud hierarchy E2E tests: 1 passed each with no skips.
• Re-ran a live Data Center metadata assertion: created, updated, and author were populated.
• Re-ran generated tool documentation consistency: all 98 tools and 25 toolsets are current.
• Confirmed the PR head remains mergeable with current main; all required CI checks are green.

Checklist

• Code follows project style guidelines (linting passes).
• Tests added/updated for changes.
• All relevant tests pass locally.
• Documentation updated (if needed).

[mcp-atlassian-maintainer]

Merged, @Cchena1. The paired regression tests for history-backed metadata and the version-only fallback made this fix easy to verify.