[#140] Add initial JMH benchmark baseline for agentscope-core

[Alexxigang]

Summary

• adds a first-phase JMH benchmark harness in �gentscope-core
• covers three core baseline paths: single ReActAgent calls, sequential pipeline orchestration, and MsgHub lifecycle/broadcast behavior
• wires Maven execution so contributors can compile and run focused benchmarks from the reactor build

Why this change

Issue #140 asks for a maintainable performance benchmark baseline. Implementing the entire matrix in one pass would be heavy and hard to review, so this PR focuses on a concrete phase 1 baseline in the core module:

• it establishes the JMH dependency and annotation-processing setup
• it adds reusable benchmark fixtures with mock-model-backed scenarios
• it makes future benchmark expansion straightforward without introducing a new top-level module yet

Validation

• mvn -pl agentscope-core -am spotless:apply
• mvn -pl agentscope-core -am -DskipTests test-compile
• mvn -pl agentscope-core -am -rf :agentscope-core -DskipTests -Dbenchmark.include=.ReActAgentBenchmark. -Dbenchmark.warmupIterations=1 -Dbenchmark.measurementIterations=1 -Dbenchmark.forks=1 test-compile exec:exec@run-benchmarks

Scope note

This PR partially addresses #140 by delivering the initial benchmark harness and first baseline scenarios. Additional benchmark cases can be layered on top of this harness in follow-up PRs.

AI assistance

This PR was prepared with AI-assisted tooling, with implementation, validation, and final review performed locally before submission.

[codecov]

Codecov Report

❌ Patch coverage is 76.92308% with 9 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
...io/agentscope/core/session/mysql/MysqlSession.java	76.92%	9 Missing ⚠️

📢 Thoughts on this report? Let us know!