fix(agent): flush pending memory retains before compaction boundary

[liuhao1024]

What does this PR do?

Context compression triggers on_session_switch() without first flushing pending memory retains. When Hindsight's turn buffer is cleared, any sync_all() dispatched after the last successful retain but before compaction fires are silently lost at the compaction boundary.

The fix adds flush_pending(timeout=5.0) before the on_session_switch call, using the same try/except pattern as the existing memory manager calls. This ensures pending sync/prefetch work completes before the buffer is cleared.

Related Issue

Fixes NousResearch#64315

Type of Change

• 🐛 Bug fix (non-breaking change that fixes an issue)
• ✨ New feature (non-breaking change that adds functionality)
• 🔒 Security fix
• 📝 Documentation update
• ✅ Tests (adding or improving test coverage)
• ♻️ Refactor (no behavior change)
• 🎯 New skill (bundled or hub)

Changes Made

• agent/conversation_compression.py: Add flush_pending(timeout=5.0) call before on_session_switch() at compaction boundary (10 lines)
• tests/agent/test_compression_memory_flush.py: Add regression test that verifies flush_pending is called with timeout=5.0 before on_session_switch during compression

How to Test

1. Run the new regression test:

   python -m pytest tests/agent/test_compression_memory_flush.py -v

   Observed result: test passes

2. Run existing compression tests to ensure no regression:

   python -m pytest tests/agent/test_compression_concurrent_fork.py -v

   Observed result: all tests pass

3. (Optional) Manual verification with Hindsight provider:

   • Configure memory.provider: hindsight in config.yaml
   • Set compression.threshold: 0.75 (default)
   • Run a long session that triggers compression
   • Check Hindsight container logs for batch_retain entries
   • Observed result: retain activity continues across compaction boundary with no gaps

Checklist

Code

• I've read the Contributing Guide
• My commit messages follow Conventional Commits (fix(scope):, feat(scope):, etc.)
• I searched for existing PRs to make sure this isn't a duplicate
• My PR contains only changes related to this fix/feature (no unrelated commits)
• I've run pytest tests/ -q and all tests pass
• I've added tests for my changes (required for bug fixes, strongly encouraged for features)
• I've tested on my platform: macOS 15.2

Documentation & Housekeeping

• I've updated relevant documentation (README, docs/, docstrings) — or N/A
• I've updated cli-config.yaml.example if I added/changed config keys — or N/A
• I've updated CONTRIBUTING.md or AGENTS.md if I changed architecture or workflows — or N/A
• I've considered cross-platform impact (Windows, macOS) per the compatibility guide — or N/A
• I've updated tool descriptions/schemas if I changed tool behavior — or N/A

For New Skills

• This skill is broadly useful to most users (if bundled) — see Contributing Guide
• SKILL.md follows the standard format (frontmatter, trigger conditions, steps, pitfalls)
• No external dependencies that aren't already available (prefer stdlib, curl, existing Hermes tools)
• I've tested the skill end-to-end: hermes --toolsets skills -q "Use the X skill to do Y"

Screenshots / Logs