feat(bench): migrate benchmarks to a GQT fixture and workload catalog

.github/workflows/gq-logic-tests.yml

@@ -183,6 +183,12 @@ jobs:
           RUSTFLAGS: ""
         run: cargo test -p omnigraph-gqt --locked --timings --lib --test runner_dispatch -- --nocapture
 
+      - name: Test ordinary GQT core execution
+        if: needs.classify_changes.outputs.run_gqt == 'true' && matrix.mode == 'ordinary'
+        env:
+          RUSTFLAGS: ""
+        run: cargo test -p omnigraph-gqt-core --locked --timings --lib -- --nocapture
+
       # A case's `at:` name arms a seam in the seam guard, so a cases-only
       # diff can change its verdict while `Test Workspace` is skipped.
       - name: Run the seam guard the case corpus arms
@@ -191,15 +197,23 @@ jobs:
           RUSTFLAGS: ""
         run: cargo test -p omnigraph-seams --locked --timings --test failpoint_names_guard
 
+      # This parity test reads a corpus case, so it must run when a
+      # cases-only diff skips `Test Workspace`.
+      - name: Test benchmark fixture parity with the GQT corpus
+        if: needs.classify_changes.outputs.run_gqt == 'true' && matrix.mode == 'ordinary'
+        env:
+          RUSTFLAGS: ""
+        run: cargo test -p omnigraph-bench --locked --timings --lib branch_merge::tests::generated_dataset_matches_the_legacy_fixture_and_reversible_history -- --exact --nocapture
+
       # The workspace .cargo/config.toml enables the seeded Tokio runtime
       # from any directory; these steps run from the repo root.
       - name: Run complete GQ logic tests with DST
         if: needs.classify_changes.outputs.run_gqt == 'true' && matrix.mode == 'dst'
-        run: cargo test -p omnigraph-gqt --locked --timings -- --nocapture
+        run: cargo test -p omnigraph-gqt -p omnigraph-gqt-core --locked --timings -- --nocapture
 
       - name: Clippy GQT with DST
         if: needs.classify_changes.outputs.run_gqt == 'true' && matrix.mode == 'dst-clippy'
-        run: cargo clippy -p omnigraph-gqt --all-targets --locked --timings -- -D warnings -W clippy::dbg_macro
+        run: cargo clippy -p omnigraph-gqt -p omnigraph-gqt-core --all-targets --locked --timings -- -D warnings -W clippy::dbg_macro
 
       - name: Retain GQT invocation reports
         if: always() && needs.classify_changes.outputs.run_gqt == 'true' && matrix.mode != 'dst-clippy'

.github/workflows/gqt-slow-nightly.yml

@@ -47,6 +47,30 @@ jobs:
       - name: Run the slow cases
         run: cargo run -p omnigraph-gqt --locked --bin omnigraph-gqt -- crates/omnigraph-gqt/cases_slow
 
+      - name: Validate full benchmark datasets
+        run: |
+          cargo run -p omnigraph-gqt --locked --bin omnigraph-gqt -- \
+            benchmarks/fixtures/branch_merge_d50.gqt \
+            benchmarks/fixtures/equal_current_narrow_one_batch.gqt \
+            benchmarks/fixtures/equal_current_wide_four_batches.gqt \
+            benchmarks/fixtures/equal_current_wide_one_batch.gqt \
+            benchmarks/fixtures/hot_table_idle_base.gqt \
+            benchmarks/fixtures/hot_table_idle_branches_4.gqt \
+            benchmarks/fixtures/hot_table_idle_branches_4_aged.gqt \
+            benchmarks/fixtures/hot_table_idle_tables_4.gqt \
+            benchmarks/fixtures/identity_lifecycle.gqt \
+            benchmarks/fixtures/nearest_ranks_updated_embedding.gqt \
+            benchmarks/fixtures/nearest_skips_deleted_rows_and_fills_limit.gqt \
+            benchmarks/fixtures/parallel_tables.gqt \
+            benchmarks/fixtures/repeated_deletion_recreation_branches.gqt \
+            benchmarks/fixtures/repeated_deletion_recreation_rows.gqt \
+            benchmarks/fixtures/threshold_crossings_after.gqt \
+            benchmarks/fixtures/threshold_crossings_at.gqt \
+            benchmarks/fixtures/threshold_crossings_before.gqt \
+            benchmarks/fixtures/threshold_crossings_production_control.gqt \
+            benchmarks/fixtures/tiny_graph.gqt \
+            benchmarks/fixtures/very_long_history_32.gqt
+
       - name: Retain GQT invocation reports
         if: always()
         uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1

Cargo.toml

@@ -18,6 +18,7 @@ members = [
     "crates/omnigraph-server",
     "crates/omnigraph-dst",
     "crates/omnigraph-gqt",
+    "crates/omnigraph-gqt-core",
     "tools/omnigraph-vocabulary-guard",
 ]
 default-members = [
@@ -111,6 +112,9 @@ uuid = { version = "1", features = ["v4"] }
 hmac = "0.12"
 httpdate = "1"
 
+[workspace.lints.rust]
+unexpected_cfgs = { level = "warn", check-cfg = ["cfg(tokio_unstable)"] }
+
 [workspace.lints.clippy]
 # Nested guards often read clearer than one merged condition (Polars allows this too).
 collapsible_if = "allow"