Add GenAI memory span instrumentation

[nagkumar91]

Summary

Implements LangChain retriever memory-search instrumentation aligned with the
GenAI memory semantic-convention proposal in
open-telemetry/semantic-conventions#3250.

What changed

• Added retriever memory-search instrumentation in
  opentelemetry-instrumentation-langchain:
  • emits search_memory operation telemetry for memory retrievers
  • records memory store/query/result-count attributes
  • handles retriever start/end/error lifecycle via the shared GenAI handler
• Added/updated LangChain unit tests and changelog entry.

Related PRs

• Semantic conventions proposal: gen-ai: semantic conventions for memory operations semantic-conventions#3250
• Python contrib implementation: Add GenAI memory span instrumentation #4215

How has this been tested?

• pytest -q instrumentation-genai/opentelemetry-instrumentation-langchain/tests/test_callback_handler_memory.py

Type of change

• New feature (non-breaking change which adds functionality)
• Bug fix (non-breaking change which fixes an issue)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

Does This PR Require a Core Repo Change?

• Yes. - Link to PR:
• No.

Checklist

• Followed the style guidelines of this project
• Changelogs have been updated
• Unit tests have been added
• Documentation has been updated

[nagkumar91]

Spec update: gen_ai.memory.type removed

FYI — we've removed gen_ai.memory.type from the semantic conventions spec (open-telemetry/semantic-conventions#3250, commit e2aabf7b).

Only Azure AI Foundry natively exposes memory types. Since LangChain doesn't expose memory types at the API level, this attribute should be removed from the instrumentation code in this PR.

Please update to stop emitting gen_ai.memory.type on memory spans.