fix(runtime): recapture late-bound DDSes during attach

[shlevari]

Description

Fixes AB#76716, where a remote client can fail with Channel context not found after a detached container is attached.

Data stores are captured sequentially when producing the attach summary. Summarizing a later data store can bind a DDS owned by an earlier, already-captured data store through normal handle serialization. The DDS then becomes visible and sends ops after attach, but remote clients never received it in the earlier data store's snapshot.

This change:

• Tracks changes to each data store's bound DDS set and recaptures stale data stores until the container-wide attach summary stabilizes.
• Conservatively stabilizes older data store runtimes that do not expose the new version capability.
• Captures per-data-store summary and GC data coherently.
• Replaces the opaque missing-channel assert with a DataProcessingError containing actionable diagnostics.

The end-to-end regression uses ordinary SharedMap handles across two data stores. It fails on the previous implementation with Channel context not found and passes with this change.

Reviewer Guidance

The review process is outlined in the pull request guidelines.

The key logic is the fixed-point attach capture in ChannelCollection. Please review the versioned path, the fingerprint fallback for legacy runtimes, and the producer-side stabilization guard.

[github-actions]

Hi! Thank you for opening this PR. Want me to review it?

Based on the diff (2060 lines, 13 files), I've queued these reviewers:

• Correctness — logic errors, race conditions, lifecycle issues
• Security — vulnerabilities, secret exposure, injection
• API Compatibility — breaking changes, release tags, type design
• Performance — algorithmic regressions, memory leaks
• Testing — coverage gaps, hollow tests

How this works

• Adjust the reviewer set by ticking/unticking boxes above. Reviewer toggles alone don't trigger anything.

• Tick Start review below to dispatch the review fleet.

• After review finishes, tick Start review again to request another run — it auto-resets after each dispatch.

• This comment updates as new commits land; your reviewer selections are preserved.

• Start review