Fixed formatting and added a comment explaining why we have a dialect…

…-specific test here.

Signed-off-by: Jason Smart <jlsmart@meta.com>

test/src/databases/snowflake/snowflake.spec.ts

@@ -7,13 +7,17 @@
 
 /* eslint-disable no-console */
 
-import {RuntimeList} from '../../runtimes';
-import {describeIfDatabaseAvailable} from '../../util';
+import { RuntimeList } from '../../runtimes';
+import { describeIfDatabaseAvailable } from '../../util';
 import '../../util/db-jest-matchers';
 
 const [describe, databases] = describeIfDatabaseAvailable(['snowflake']);
 const runtimes = new RuntimeList(databases);
-
+/**
+Custom tests for Snowflake.  The HyperLogLog algorithm is approximate, and different database
+implementations can return slighlty different results.  Thus we implement per-dialect unit tests
+for these databases.
+*/
 describe.each(runtimes.runtimeList)(
   'Snowflake dialect functions - %s',
 
@@ -27,7 +31,7 @@ describe.each(runtimes.runtimeList)(
       } -> {
           aggregate: code_count is hll_estimate(hll_combine(code_hll))
       }
-      `).malloyResultMatches(runtime, {code_count: 19812});
+      `).malloyResultMatches(runtime, { code_count: 19812 });
     });
   }
 );