Check initialization state while holding the data-store lock

[rudycelekli]

A delayed initializer can pass the existing-state preflight check, then replace a newer release installed by another initializer before lock acquisition. Move that precondition inside the existing exclusive update lock.

The deterministic interleaving regression fails before the fix and passes afterward. It verifies that the winning version remains active, the delayed call returns already_initialized, and the older bundle is never requested. The focused data-sync suite passes 21 tests. Removing the under-lock guard deliberately makes the regression fail.

Full clean-checkout validation

Exact signed head: f63234203477edb3741a7ec88a0570a0bee6b27c. Recursive paper submodule initialized at its repository-pinned commit. All six required commands pass:

• ruff check .
• ruff format --check .
• benchmark-radar normalize-catalog
• benchmark-radar classify
• benchmark-radar build-data-release
• pytest -q: 1466 passed, no skipped tests

The rebuilt corpus retains 3,212 records, five sources, and 67 snapshots. Generated outputs are not committed. Signed commit with DCO sign-off. GitHub checks are separate from these local results.

[rudycelekli]

Agent-authored PR annotation: 330226 GPT.

[ktwu01]

Closing as not planned. One author now has a cap of 10 open issues + PRs, and this account had 56 open PRs and 11 open issues, mostly paired with each other. We merged the three changes that fix silent data loss in the corpus (#723, #774, and #773 once rebased) and are closing the rest so review stays possible.

This is not a verdict that the change is wrong. If you think this one matters, reopen it later when you are under the cap, one at a time, with a reproduction from a real user or caller rather than a hypothetical edge case.

不是否定这个改动本身。为保证评审能跟上，同一作者最多同时保留 10 个未关闭的 issue/PR；该账号曾有 56 个未关闭 PR 和 11 个 issue。已合并修复语料静默丢失的三个改动，其余先关闭。若认为本条重要，请在低于上限后逐个重新打开，并附真实用户或调用方的复现。

---

Generated by Claude Code