feat(langchain): Support v1

[sentrivana]

Description

Adapt to import changes in new langchain 1.x

Issues

Closes #4735

Reminders

• Please add tests to validate your changes, and lint your code using tox -e linters.
• Add GH Issue ID & Linear ID (if applicable)
• PR title should use conventional commit style (feat:, fix:, ref:, meta:)
• For external contributors: CONTRIBUTING.md, Sentry SDK development docs, Discord community

[codecov]

❌ 2 Tests Failed:

Tests completed	Failed	Passed	Skipped
27447	2	27445	2004

View the full list of 2 ❄️ flaky test(s)

tests.profiler.test_continuous_profiler::test_continuous_profiler_auto_start_and_manual_stop[non-experiment-start_profile_session/stop_profile_session (deprecated)-gevent]

Flake rate in main: 61.70% (Passed 18 times, Failed 29 times)

Stack Traces | 0.555s run time

tests/profiler/test_continuous_profiler.py:291: in test_continuous_profiler_auto_start_and_manual_stop
    assert_single_transaction_with_profile_chunks(envelopes, thread)
tests/profiler/test_continuous_profiler.py:147: in assert_single_transaction_with_profile_chunks
    assert len(items["profile_chunk"]) > 0
E   assert 0 > 0
E    +  where 0 = len([])

tests.profiler.test_continuous_profiler::test_continuous_profiler_auto_start_and_manual_stop[non-experiment-start_profiler/stop_profiler-gevent]

Flake rate in main: 61.70% (Passed 18 times, Failed 29 times)

Stack Traces | 0.548s run time

tests/profiler/test_continuous_profiler.py:291: in test_continuous_profiler_auto_start_and_manual_stop
    assert_single_transaction_with_profile_chunks(envelopes, thread)
tests/profiler/test_continuous_profiler.py:147: in assert_single_transaction_with_profile_chunks
    assert len(items["profile_chunk"]) > 0
E   assert 0 > 0
E    +  where 0 = len([])

To view more test analytics, go to the Test Analytics Dashboard
_{📋 Got 3 mins? Take this short survey to help us improve Test Analytics.}

[cursor]

Bug: AgentExecutor Import Overwrite Bug

The AgentExecutor import logic has two sequential try/except blocks. The second block unconditionally executes, causing langchain.agents.AgentExecutor to overwrite a successful langchain_classic.agents.AgentExecutor import (intended for v1+). This also means a successful langchain_classic import is lost if the langchain import fails, incorrectly setting AgentExecutor to None.