perf(python): add O2 and O4 operator benchmarks (#98)

## Summary
- Add **O2** benchmark: fractional operator with 8 weighted buckets
(multi-variant experiment)
- Add **O4** benchmark: semver range comparison using caret (`^`)
operator
- Update O1/O3 docstrings to reference their benchmark IDs for clarity

Closes #88

## Test plan
- [x] `pytest benchmarks/ -v --benchmark-disable` passes for both new
tests
- [x] Existing benchmarks unaffected (32 passed, 4 pre-existing failures
from missing `json_logic` module)
- [x] `cargo fmt --check && cargo clippy -- -D warnings` passes

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

Author: aepfli

python/benchmarks/conftest.py

@@ -101,7 +101,35 @@ def _build_flag_config():
                     ]
                 },
             },
-            # Semver flag
+            # Fractional bucketing flag with 8 weighted buckets (O2)
+            "fractional-8-flag": {
+                "state": "ENABLED",
+                "variants": {
+                    "v1": "v1",
+                    "v2": "v2",
+                    "v3": "v3",
+                    "v4": "v4",
+                    "v5": "v5",
+                    "v6": "v6",
+                    "v7": "v7",
+                    "v8": "v8",
+                },
+                "defaultVariant": "v1",
+                "targeting": {
+                    "fractional": [
+                        {"var": "targetingKey"},
+                        ["v1", 12],
+                        ["v2", 13],
+                        ["v3", 12],
+                        ["v4", 13],
+                        ["v5", 12],
+                        ["v6", 13],
+                        ["v7", 12],
+                        ["v8", 13],
+                    ]
+                },
+            },
+            # Semver flag (equality, O3)
             "semver-flag": {
                 "state": "ENABLED",
                 "variants": {"new-ui": True, "old-ui": False},
@@ -114,6 +142,19 @@ def _build_flag_config():
                     ]
                 },
             },
+            # Semver range flag with caret operator (O4)
+            "semver-range-flag": {
+                "state": "ENABLED",
+                "variants": {"on": True, "off": False},
+                "defaultVariant": "off",
+                "targeting": {
+                    "if": [
+                        {"sem_ver": [{"var": "version"}, "^", "1.2.0"]},
+                        "on",
+                        "off",
+                    ]
+                },
+            },
             # starts_with flag
             "starts-with-flag": {
                 "state": "ENABLED",

python/benchmarks/test_benchmarks.py

@@ -124,7 +124,7 @@ class TestCustomOperatorBenchmarks:
     """Benchmarks for flagd custom JSON Logic operators."""
 
     def test_bench_fractional_operator(self, benchmark, evaluator):
-        """Fractional bucketing operator (MurmurHash3-based)."""
+        """Fractional bucketing operator with 3 buckets (O1)."""
         ctx = {"targetingKey": "user-abc-123"}
         result = benchmark(evaluator.evaluate_string, "fractional-flag", ctx, "fallback")
         assert result in [
@@ -133,12 +133,24 @@ def test_bench_fractional_operator(self, benchmark, evaluator):
             "treatment-b-experience",
         ]
 
+    def test_bench_fractional_8_buckets(self, benchmark, evaluator):
+        """Fractional bucketing with 8 weighted buckets — multi-variant experiment (O2)."""
+        ctx = {"targetingKey": "user-abc-123"}
+        result = benchmark(evaluator.evaluate_string, "fractional-8-flag", ctx, "fallback")
+        assert result in ["v1", "v2", "v3", "v4", "v5", "v6", "v7", "v8"]
+
     def test_bench_semver_operator(self, benchmark, evaluator):
-        """Semantic version comparison operator."""
+        """Semantic version equality comparison (O3)."""
         ctx = {"appVersion": "2.5.1"}
         result = benchmark(evaluator.evaluate_bool, "semver-flag", ctx, False)
         assert result is True
 
+    def test_bench_semver_range_operator(self, benchmark, evaluator):
+        """Semantic version range comparison with caret operator (O4)."""
+        ctx = {"version": "1.5.3"}
+        result = benchmark(evaluator.evaluate_bool, "semver-range-flag", ctx, False)
+        assert result is True
+
     def test_bench_starts_with_operator(self, benchmark, evaluator):
         """String prefix matching operator."""
         ctx = {"email": "admin@example.com"}