Harden new tsid hashing transport version

[dnhatn]

This commit introduced a transport version for the new tsid hash, which was missing in #98023. At the time, the latest version was ASYNC_SEARCH_STATUS_SUPPORTS_KEEP_ALIVE = def(8_582_00_0). Therefore, I introduced a new version: TransportVersion TIME_SERIES_ID_HASHING = def(8_582_10_0). This change aims to minimize potential issues during upgrades from 8.13.0 to later versions in serverless.

elasticsearch/server/src/main/java/org/elasticsearch/TransportVersions.java

Line 169 in bdd3a4f

public static final TransportVersion ASYNC_SEARCH_STATUS_SUPPORTS_KEEP_ALIVE = def(8_582_00_0);

Relates #98023
Relates #106878

[elasticsearchmachine]

Pinging @elastic/es-storage-engine (Team:StorageEngine)

[ldematte]

Overall seems OK, after conversation on slack I understand this is a refinement on the previous PR.
I'm sure you double-checked this is the correct version. There is one change to be made though, the "part" you used is not for patching.

[ldematte]

You should not use that for a patch version -- those are reserved for serverless.
From the docs a few lines below:

     * M_NNN_SS_P
     *
     * M - The major version of Elasticsearch
     * NNN - The server version part
     * SS - The serverless version part. It should always be 00 here, it is used by serverless only.
     * P - The patch version part

[salvatore-campagna]

My understanding is that this is not really required if no customer in Serverless deployments has ever used downsampling. If they did they would have had a corrupted cluster state and we would have seen the issue in Serverless. We also know that all Serverless deployments are already on 8.13 and we didn't see any issue there which means that no one of our Serverless customers has been using downsampling before the latest Serverless release. Any customer using version 8.13, either on-prem or on Serverless, is already using the new cluster state and they would not experience any issue.

Am I wrong?

[kkrik-es]

Since there's no tangible benefit from submitting it, I'd say we skip.