fix(skill): scope fresh semantic chunks before update merges - #4123

Compared merge-base b57cf9108ea3ae571f3e6b3f5dff7de900750127 (where this PR branched from the base branch) against head b0358e645cc85aeb82c6a3b609e4b8002108840d. Changes on the base branch since then are not attributed to this PR.

PR-changed functions: 12/21 verified (0 proven, 12 may-equivalent, 0 distinguished) · 9 not verified (7 vacuous, 2 unsupported)

• graphify/build\.py · build: unsupported (vacuous: never exercised, not verified) — no capturable inputs from the test suite; property tier: not verifiable: all 9 sampled inputs raised on both versions — the function never executed, so 'no divergence' would be vacuous (mostly AttributeError — names the real obstacle, not…
• graphify/build\.py · build_from_json: unsupported (vacuous: never exercised, not verified) — no capturable inputs from the test suite; property tier: not verifiable: all 11 sampled inputs raised on both versions — the function never executed, so 'no divergence' would be vacuous
• graphify/cli\.py · dispatch_command: unsupported (vacuous: never exercised, not verified) — no capturable inputs from the test suite; property tier: not verifiable: all 40 sampled inputs raised on both versions — the function never executed, so 'no divergence' would be vacuous (mostly IndexError — names the real obstacle, not a…
• graphify/detect\.py · detect_incremental: unsupported — no capturable inputs from the test suite; property tier: nondeterministic output (both versions disagreed with themselves on the witness) — not a behaviour change; a function whose output is not a function of its inputs (random / time /…
• graphify/exporters/graphdb\.py · push_to_falkordb: unsupported (vacuous: never exercised, not verified) — no capturable inputs from the test suite; property tier: not verifiable: all 264 sampled inputs raised on both versions — the function never executed, so 'no divergence' would be vacuous (mostly ImportError — names the real obstacle, not…
• graphify/exporters/graphdb\.py · push_to_neo4j: unsupported (vacuous: never exercised, not verified) — no capturable inputs from the test suite; property tier: not verifiable: all 261 sampled inputs raised on both versions — the function never executed, so 'no divergence' would be vacuous (mostly ImportError — names the real obstacle, not…
• graphify/extractors/engine\.py · _extract_generic: unsupported — no capturable inputs from the test suite; property tier: parameter `config` is annotated `LanguageConfig` — outside the synthesizable primitive/collection set
• graphify/extractors/engine\.py · _js_extra_walk: unsupported (vacuous: never exercised, not verified) — no capturable inputs from the test suite; property tier: not verifiable: all 264 sampled inputs raised on both versions — the function never executed, so 'no divergence' would be vacuous (mostly AttributeError — names the real obstacle, n…
• graphify/paths\.py · load_node_link_graph: unsupported (vacuous: never exercised, not verified) — no capturable inputs from the test suite; property tier: not verifiable: all 52 sampled inputs raised on both versions — the function never executed, so 'no divergence' would be vacuous (mostly TypeError — names the real obstacle, not a s…

Backend availability notices:

• 3 changed test function(s) excluded from verification (a test is the oracle, not the subject)
• none of the 0 changed id(s) matched a code node in the graph (0 known) — 24 changed function(s) recovered directly from the git diff instead (graph may be stale/unsynced; consider a rebuild)

Evidence certificates (JSON / Markdown): open in Graphify (sign-in required).

Formal verification is advisory: equivalent is a proof over a bounded sound subset; may_equivalent is sampled, not proven; unsupported and error are honest abstentions. A vacuous check (it never exercised the function) is counted as not verified. Only functions this PR changed are counted; context functions are listed separately.