[EXPORTER] Fix incorrect doc comment about SessionState::TimedOut in the Elasticsearch exporter

[om7057]

Fixes #4604 (doc comment only; the enum question needs SIG discussion, see below)

Changes

ResponseHandler::waitForResponse()'s doc comment in the Elasticsearch log exporter claims a request timeout (set via SetTimeoutMs()) "arrives here as a TimedOut session event." That's incorrect: no HttpClient implementation in this repo dispatches SessionState::TimedOut. The curl client maps CURLE_OPERATION_TIMEDOUT to SendFailed, indistinguishable from any other transport failure, which the comment now says.

This PR originally also removed the unreachable TimedOut enumerator and its four dead branches (Elasticsearch and OTLP HTTP exporters). A reviewer correctly pointed out that's not something to do directly: SessionState is part of an installed public header and the event contract for any out-of-tree HttpClient implementation, so removing an enumerator needs to go through docs/deprecation-process.md (SIG discussion, a [DEPRECATION] issue, a DEPRECATED entry, a full deprecation cycle before removal). Issue #4604 is also still needs-triage, so it hasn't been accepted for a specific direction yet.

Scoped this PR down to just the doc comment fix, which needs no process to correct. Left the actual enum question (dispatch TimedOut for real vs. deprecate-and-remove it) for the SIG to decide on the issue.

Ran the full test suite locally, all passing.

• CHANGELOG.md updated for non-trivial changes

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 86.66%. Comparing base (754928d) to head (b3151b9).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #4647   +/-   ##
=======================================
  Coverage   86.66%   86.66%           
=======================================
  Files         525      525           
  Lines       20481    20481           
=======================================
  Hits        17748    17748           
  Misses       2733     2733           

Files with missing lines	Coverage Δ
...orters/elasticsearch/src/es_log_record_exporter.cc	47.73% <ø> (ø)

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[nikhilbhatia08]

Thanks for the investigation — the diagnosis is right, but I don't think we can remove the enumerator. SessionState is in an installed public header and is the event contract for any HttpClient implementation, including the out-of-tree ones we explicitly support. Removing it breaks their build, so it would need a DEPRECATED entry and a deprecation cycle per docs/deprecation-process.md.

Also, #4604 is still needs-triage that needs a SIG discussion.

[om7057]

@nikhilbhatia08 Understood and thanks.
I checked docs/deprecation-process.md and you're right that removing a public enumerator needs SIG discussion and a proper deprecation cycle, not a direct removal.
Also I missed that #4604 is still needs-triage.

I have scoped this down to just the doc comment fix, which was simply wrong about the event and needs no process to correct.
Reverted the enum removal and the four dead branches back to how they were. Left the actual TimedOut question (dispatch it for real vs. deprecate-and-remove) for the SIG to decide on the issue.