ci(workflows): two more comments cite closed issues as live justification (#5263, #14371) #15346

What

Building the closed-reference guard for #15333 AC4 surfaced two more instances
of the same defect #15252 and #15333 AC3 already named (a comment citing a now
-closed issue as the reason a gap is acceptable):

• code-quality.yml:258 — "doesn't block on the pre-existing violations
  tracked by discovery(slm/infra): 59 pre-existing datetime.utcnow().isoformat() / + Z sites in autobot-slm-backend, autobot-slm-agent, autobot-infrastructure #5263 (22 in 10 files as of fix(ci): code-quality steps compare against a base commit absent from the shallow clone #13880 ...)" and
  code-quality.yml:260 — "After discovery(slm/infra): 59 pre-existing datetime.utcnow().isoformat() / + Z sites in autobot-slm-backend, autobot-slm-agent, autobot-infrastructure #5263 closes, swap to full-scan mode (no
  argv)." #5263 is CLOSED. Per the comment's own stated trigger, the
  "Block datetime.utcnow().isoformat() regressions" step should already have
  been swapped from changed-files-only to full-scan mode, and has not been.
• ssot-coverage.yml:172 — the PR-comment body template says "Known backlog
  in pipeline-scripts/hardcoded_values_baseline.txt is suppressed and
  tracked in tech-debt: three separate hardcoded-value detectors have drifted (pipeline-scripts, autobot-infrastructure, pre-commit hook) #14371." #14371 is CLOSED ("tech-debt: three separate
  hardcoded-value detectors have drifted").

Why not fixed in this PR

Both require a judgment call this PR's scope (#15332/#15333/#15252) doesn't
extend to: whether #5263 closing means the datetime-regression guard should
genuinely switch to full-scan mode now (and what that scan would find), and
whether the hardcoded_values_baseline.txt backlog behind #14371 is still
accurate. Fixing the citation without checking the underlying claim would
repeat exactly the mistake #15252/#15333 already documented.

The new repo_tests guard added in #15333 (closed-reference check) carries an
explicit, reasoned exemption for these two citations pointing at this issue,
so the guard does not silently pass on them — the exemption is visible and
temporary, not a permanent allowlist.

Acceptance criteria

• code-quality.yml:258-260 — decide whether to re-measure #5263's
  violation count and swap to full-scan mode, or re-justify staying in
  changed-files-only mode with current evidence
• ssot-coverage.yml:172 — repoint #14371 at whatever issue currently
  tracks the hardcoded-values baseline backlog, or re-justify in place
• Remove both entries from the guard's exemption list once fixed