feat: implement high-precision performance tracing in executor

[vansham]

I have added a high-precision execution timer in executor/src/lib.rs using std::time::Instant. This allows the GenVM to log exact execution time in microseconds for every Intelligent Contract call. This is a critical tool for developers to benchmark and optimize their contracts.

Summary by CodeRabbit

• New Features
  • Execution-time metrics recorded per run and surfaced in performance traces and logs to help spot slow contracts.
  • Configuration now includes runner and registry paths and performs pre-run validation of required addresses and directories, with clear errors on misconfiguration.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[CLAassistant]

All committers have signed the CLA.

[coderabbitai]

📝 Walkthrough

Walkthrough

Adds execution-time instrumentation to the executor and expands configuration validation: introduces execution_time_us: u128 to Metrics, uses std::time::Instant to time VM runs and populate/log execution duration, and adds new config fields (runners_dir, registry_dir), derives, and a Config::validate() method called from supervisor creation.

Changes

Cohort / File(s)	Summary
Performance timing & metrics executor/src/lib.rs	Added pub execution_time_us: u128 to Metrics; imported std::time::Instant; instrumented create_supervisor to call config.validate(); instrumented run_with_impl and run_with to record start/end times, set metrics.execution_time_us, and emit performance trace logs with contract address and duration.
Configuration and validation executor/src/config.rs	Added Serialize, Debug, Clone derives to Module, Modules, and Config; added public fields runners_dir: String and registry_dir: String; introduced pub fn validate(&self) -> Result<()> that checks LLM/Web addresses and cache_dir/runners_dir presence and logs validation success. Updated imports for Serialize, anyhow::Result, and bail.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Poem

🐇 I tapped my paw upon the clock,
I chased each microsecond tick-tock,
Metrics now gleam from start to end,
Configs checked before we send,
✨⏱️ — a rabbit's hop to ship and stock.

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 33.33% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The PR title accurately reflects the main change: adding execution time tracking in microseconds to the executor's Metrics struct with performance logging.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

• 📝 Generate docstrings

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

executor/src/lib.rs (1)

98-201: Timing is skipped on early error paths
The vm.run().await? and the VM-error early return bypass the log block, so failed calls won’t emit timing even though the intent is “every call.” Consider logging before propagating errors or using a small guard to ensure the log fires on all exits.

🛠️ Minimal fix to log even when vm.run() fails

-    let run_result = vm.run().await?;
+    let run_result = vm.run().await;

     // --- LOG PERFORMANCE METRICS ---
     let exec_duration = start_instant.elapsed().as_micros();
     log_info!(
         contract = %entry_data.message.contract_address,
         duration_us = %exec_duration;
         "Intelligent Contract execution performance trace"
     );
     // -------------------------------
+
+    let run_result = run_result?;

(You’ll still want to route the earlier VM-error return through the same logging path or a guard to cover that branch too.)

🤖 Fix all issues with AI agents

In `@executor/src/lib.rs`:
- Around line 31-33: The Metrics struct's execution_time_us field is never set
so serialized metrics stay zero; either remove the field or wire it to the
actual exec duration: locate where exec_duration is computed in run_with and
assign that value (converted to u128 microseconds) into
Metrics::execution_time_us before serializing/sending, ensuring proper
synchronization if Metrics is shared across threads, or remove execution_time_us
from the Metrics definition and any serialization/consumption sites (and keep
only the logging of exec_duration) to avoid misleading data.

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@executor/src/lib.rs`:
- Around line 98-100: The timing start (let start_instant = Instant::now()) is
only recorded on the happy path and early returns skip setting execution_time_us
and the log; introduce a small RAII/Drop guard (e.g., ExecutionTimerGuard)
created where start_instant is now, have its Drop impl compute duration, update
execution_time_us and emit the log/metrics, and ensure any references to
start_instant are replaced by or use the guard so that timing is recorded on all
exit paths (including errors and panics); attach the guard at the top of the
function scope that currently contains start_instant so it runs regardless of
how the function returns.

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@executor/src/config.rs`:
- Around line 26-45: The Config::validate method is missing the log_info! macro
import and omits validation for registry_dir; fix by adding the common import
(e.g., bring in genvm_common symbols so log_info! is available) at the top of
the file and extend the validation logic inside Config::validate to check that
self.registry_dir is not empty alongside self.cache_dir and self.runners_dir,
returning a bail! error similar to the existing messages if it is empty; keep
existing checks for modules.llm.address and modules.web.address intact.