Deduplicate Index Meta Generations when Deserializing (#65619) (#65664)

These strings are quite long individually and will be repeated
potentially up to the number of snapshots in the repository times.
Since these make up more than half of the size of the repository metadata
and are likely the same for all snapshots the savings from deduplicating them
can make up for more than half the size of `RepositoryData` easily in most real-world
cases.

Author: original-brownbear

server/src/main/java/org/elasticsearch/repositories/RepositoryData.java

@@ -608,6 +608,7 @@ private static void parseSnapshots(XContentParser parser, Map<String, SnapshotId
                                        Map<String, Version> snapshotVersions,
                                        Map<SnapshotId, Map<String, String>> indexMetaLookup) throws IOException {
         XContentParserUtils.ensureExpectedToken(XContentParser.Token.START_ARRAY, parser.nextToken(), parser);
+        final Map<String, String> stringDeduplicator = new HashMap<>();
         while (parser.nextToken() != XContentParser.Token.END_ARRAY) {
             String name = null;
             String uuid = null;
@@ -628,7 +629,7 @@ private static void parseSnapshots(XContentParser parser, Map<String, SnapshotId
                         state = SnapshotState.fromValue((byte) parser.intValue());
                         break;
                     case INDEX_METADATA_LOOKUP:
-                        metaGenerations = parser.mapStrings();
+                        metaGenerations = parser.map(HashMap::new, p -> stringDeduplicator.computeIfAbsent(p.text(), Function.identity()));
                         break;
                     case VERSION:
                         version = Version.fromString(parser.text());