HDDS-15942. Remove Pair usage from hadoop-hdds module

[navinko]

What changes were proposed in this pull request?

Removed org.apache.commons.lang3.tuple.Pair usage from hadoop-hdds/*

Please describe your PR in detail:

Removed Pair usage from hadoop-hdds and replaced it with named value classes so the code is clearer and the APIs are easier to understand. I replaced generic Pair usages with named value classes like SafeModeRuleStatus, ScmNodeAddress, PendingRead, RpcRequest, and similar helpers.
I updated the affected interfaces, implementations, and tests so everything still matches and compiles together.

What is the link to the Apache JIRA

https://issues.apache.org/jira/browse/HDDS-15942

How was this patch tested?

Updated related test cases and ran locally.
Successful CI build : https://github.com/navinko/ozone/actions/runs/31095580829

[szetszwo]

@navinko , thanks for working on this! This PR is way too big. Please file subtasks and split this into multiple PRs. A good size is ~30 KB patch.

(You may use git diff commit# to generate a patch, where commit# is the last commit before your change.)

[navinko]

@navinko , thanks for working on this! This PR is way too big. Please file subtasks and split this into multiple PRs. A good size is ~30 KB patch.

(You may use git diff to generate a patch, where the last commit before your change.)

Thank you @szetszwo for the review . Sure , I will do that .

[szetszwo]

... Sure , I will do that .

One way to do it is to fix the problem module-by-module. Of course, if the fix in a module is really small, we may combine a few modules into one PR.

[navinko]

One way to do it is to fix the problem module-by-module. Of course, if the fix in a module is really small, we may combine a few modules into one PR.

Sure @szetszwo for the clarification, I agree this is too large for one PR. I initially thought to make all changes related to haddop-hdds module, only in one PR but it became large so i wud further divides this into submodule and open the PR