refactor(duties_tracker): retain proposer-duty v2 response metadata

[petarjuki7]

Problem, Evidence, and Context

• The tracker stored whatever proposer-duty list the beacon node returned, with no completeness check. A partial or empty response became the view for its epoch, and proposer_assignment_at_slot then answered NotAssigned for validators the response simply failed to include. SIP-101 allows a schedule to establish a view only when it is complete: one duty for every slot of the epoch.
• The response metadata was discarded. The v2 endpoint's dependent_root is the fork-aware decision root (v1 returns the legacy one), which later refresh logic needs.
• Closes refactor(duties_tracker): retain proposer-duty v2 response metadata #1183. Builds on fix(duties_tracker): validate proposer assignments against complete current/next views #1182.

Change Overview

• New ProposerSchedule type with private fields, built only through from_response, which accepts a response only if it has exactly SLOTS_PER_EPOCH duties, all inside the requested epoch, each slot exactly once. Anything else is rejected with a typed error naming the defect.
• The poll switches to get_validator_duties_proposer_v2 and stores the validated schedule together with dependent_root and execution_optimistic. Neither field affects whether a schedule is accepted; they are retained for refresh decisions and diagnostics (fix(duties_tracker): authoritative complete-schedule transitions for proposer assignment verdicts #1184).
• A failed or malformed refresh leaves the previously stored schedule in place; an epoch with no prior schedule stays absent.
• Intentionally unchanged: the DutiesProvider trait and DutyAssignment semantics, per-epoch insert ordering, current+next polling, retention pruning, and no filtering by the local registry.

Risks, Trade-offs, and Mitigations

• Scope is proposer-duty storage in this crate; dependent crates compile unchanged.
• Behavior change: malformed responses no longer replace a stored view. This is the intended SIP-101 behavior, not a regression.
• Requires the beacon node to serve /eth/v2/validator/duties/proposer; the pinned Lighthouse does.

Validation

• Unit tests cover the acceptance matrix: complete out-of-order schedule accepted, metadata retained verbatim, and wrong-length, empty, out-of-epoch, and duplicate-slot responses rejected with the exact error. With the length pinned at SLOTS_PER_EPOCH and every slot in range, a missing slot can only appear as a duplicate, so that case covers both.
• cargo +nightly fmt --check, clippy --all-targets -D warnings, and cargo test -p duties_tracker (11 tests) pass.

Rollback

• Revert the commit. No config, data, or operational impact.

Blockers / Dependencies

• fix(duties_tracker): authoritative complete-schedule transitions for proposer assignment verdicts #1184 builds on the stored type introduced here.

Additional Info / Next Steps

• The poll loop itself (v2 selection; prior view kept across a failed or malformed refresh) has no direct test: the crate cannot fake a beacon-node response today. The validation logic is a plain function and fully covered; fetch-path coverage is tracked in test(duties_tracker): cover poll_beacon_proposers fetch path via a beacon-node seam #1240.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 93.28859% with 10 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (epbs@7ec3a06). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
anchor/duties_tracker/src/duties_tracker.rs	90.19%	10 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             epbs    #1241   +/-   ##
=======================================
  Coverage        ?   73.13%           
=======================================
  Files           ?      170           
  Lines           ?    34495           
  Branches        ?        0           
=======================================
  Hits            ?    25227           
  Misses          ?     9268           
  Partials        ?        0           

Flag	Coverage Δ
rust	73.13% <93.28%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[shane-moore]

very nice!

[mergify]

Queued — the merge queue status continues in this comment ↓.

[mergify]

Merge Queue Status

• ✅ Entered queue — 2026-08-12 06:12 UTC · Rule: default · triggered by @petarjuki7 with the merge queue checkbox
• ✅ Checks passed · on draft merge queue: checking epbs (7ec3a06) and [#1241 + #1203] together #1242
• ✅ Merged — 2026-08-12 06:40 UTC · at 4faad42dfcbc986a45fee92d9fd9ea85ec7b22ba · squash

This pull request spent 28 minutes 42 seconds in the queue, including 26 minutes 44 seconds running CI.

Required conditions to merge

• check-success=cli-reference-check
• check-success=run-local-testnet
• check-success=test-suite-success