Fix flaky test #109978 (#110245)

CCS tests could split the vectors over any number of shards. Through
empirical testing, I determined this commits values work to provide the
expected order, even if they are not all part of the same shard. 

quantization can have weird behaviors when there are uniform values,
just like this test does.

closes #109978

muted-tests.yml

@@ -97,10 +97,6 @@ tests:
 - class: "org.elasticsearch.xpack.searchablesnapshots.FrozenSearchableSnapshotsIntegTests"
   issue: "https://github.com/elastic/elasticsearch/issues/110408"
   method: "testCreateAndRestorePartialSearchableSnapshot"
-- class: org.elasticsearch.test.rest.yaml.CcsCommonYamlTestSuiteIT
-  method: test {p0=search.vectors/41_knn_search_half_byte_quantized/Test create, merge,
-    and search cosine}
-  issue: https://github.com/elastic/elasticsearch/issues/109978
 - class: "org.elasticsearch.xpack.esql.qa.mixed.MixedClusterEsqlSpecIT"
   issue: "https://github.com/elastic/elasticsearch/issues/110591"
 

rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search.vectors/41_knn_search_half_byte_quantized.yml

@@ -428,7 +428,7 @@ setup:
         index: hnsw_byte_quantized_merge_cosine
         id: "1"
         body:
-          embedding: [1.0, 1.0, 1.0, 1.0]
+          embedding: [0.5, 0.5, 0.5, 0.5, 0.5, 1.0]
 
   # Flush in order to provoke a merge later
   - do:
@@ -439,7 +439,7 @@ setup:
         index: hnsw_byte_quantized_merge_cosine
         id: "2"
         body:
-          embedding: [1.0, 1.0, 1.0, 2.0]
+          embedding: [0.0, 0.0, 0.0, 1.0, 1.0, 0.5]
 
   # Flush in order to provoke a merge later
   - do:
@@ -450,7 +450,7 @@ setup:
         index: hnsw_byte_quantized_merge_cosine
         id: "3"
         body:
-          embedding: [1.0, 1.0, 1.0, 3.0]
+          embedding: [0.0, 0.0, 0.0, 0.0, 0.0, 10.5]
 
   - do:
       indices.forcemerge:
@@ -468,7 +468,7 @@ setup:
           query:
             knn:
               field: embedding
-              query_vector: [1.0, 1.0, 1.0, 1.0]
+              query_vector: [1.0, 1.0, 1.0, 1.0, 1.0, 1.0]
               num_candidates: 10
 
   - length: { hits.hits: 3 }