fix(eap-spans): formulas should not have labels within them (#89477)

It appears that if we use the same label names between multiple
expressions, even within child labels (labelA and labelC in the example
below), snuba will treat those expressions as the same. We could
dynamically create labels to avoid duplicates, but as I can't seem to
see any benefit to "child" labels.

This PR removes all child labels within formulas.

```python
Expression(
    formula=Expression.BinaryFormula(
        left=Expression(..., label="labelA") # Label A must be completely unique across all expressions in the request
        op=divide
        right=Expression(..., label="labelB") # Label B must be completely unique across all expressions in the request
    )
    label="labelC" # Label C must be completely unique across all expressions in the request
)
```

Author: DominikB2014

src/sentry/search/eap/spans/formulas.py

@@ -49,7 +49,6 @@ def get_total_span_count(settings: ResolverSettings) -> Column:
         aggregation=AttributeAggregation(
             aggregate=Function.FUNCTION_COUNT,
             key=AttributeKey(type=AttributeKey.TYPE_DOUBLE, name="sentry.exclusive_time_ms"),
-            label="total",
             extrapolation_mode=extrapolation_mode,
         )
     )
@@ -60,9 +59,9 @@ def division(args: ResolvedArguments, _: ResolverSettings) -> Column.BinaryFormu
     divisor = cast(AttributeKey, args[1])
 
     return Column.BinaryFormula(
-        left=Column(key=dividend, label="dividend"),
+        left=Column(key=dividend),
         op=Column.BinaryFormula.OP_DIVIDE,
-        right=Column(key=divisor, label="divisor"),
+        right=Column(key=divisor),
     )
 
 
@@ -143,7 +142,6 @@ def failure_rate(_: ResolvedArguments, settings: ResolverSettings) -> Column.Bin
                         ),
                     )
                 ),
-                label="trace_status_count",
                 extrapolation_mode=extrapolation_mode,
             ),
         ),
@@ -274,7 +272,6 @@ def http_response_rate(args: ResolvedArguments, settings: ResolverSettings) -> C
                         ),
                     )
                 ),
-                label="error_request_count",
                 extrapolation_mode=extrapolation_mode,
             ),
         ),
@@ -286,7 +283,6 @@ def http_response_rate(args: ResolvedArguments, settings: ResolverSettings) -> C
                     name="sentry.status_code",
                     type=AttributeKey.TYPE_STRING,
                 ),
-                label="total_request_count",
                 extrapolation_mode=extrapolation_mode,
             ),
         ),
@@ -318,7 +314,6 @@ def trace_status_rate(args: ResolvedArguments, settings: ResolverSettings) -> Co
                         ),
                     )
                 ),
-                label="trace_status_count",
                 extrapolation_mode=extrapolation_mode,
             ),
         ),
@@ -350,7 +345,6 @@ def cache_miss_rate(_: ResolvedArguments, settings: ResolverSettings) -> Column.
                         ),
                     )
                 ),
-                label="cache_miss_count",
                 extrapolation_mode=extrapolation_mode,
             ),
         ),
@@ -362,7 +356,6 @@ def cache_miss_rate(_: ResolvedArguments, settings: ResolverSettings) -> Column.
                     name="cache.hit",
                     type=AttributeKey.TYPE_BOOLEAN,
                 ),
-                label="total_cache_count",
                 extrapolation_mode=extrapolation_mode,
             ),
         ),
@@ -386,7 +379,6 @@ def ttfd_contribution_rate(
                         value=AttributeValue(val_str="ttfd"),
                     )
                 ),
-                label="ttfd_count",
                 extrapolation_mode=extrapolation_mode,
             ),
         ),
@@ -412,7 +404,6 @@ def ttid_contribution_rate(
                         value=AttributeValue(val_str="ttid"),
                     )
                 ),
-                label="ttid_count",
                 extrapolation_mode=extrapolation_mode,
             ),
         ),

tests/snuba/api/endpoints/test_organization_events_stats_span_indexed.py

@@ -1727,9 +1727,6 @@ def test_http_response_rate(self):
         assert data["data"][1][1][0]["count"] == 0.5
         assert data["data"][2][1][0]["count"] == 0.75
 
-    @pytest.mark.xfail(
-        reason="raw sql is is not querying for 5xx's, https://github.com/getsentry/eap-planning/issues/242"
-    )
     def test_http_response_rate_multiple_series(self):
         self.store_spans(
             [