🔊 Add metrics to log MM embedding time

[gkumbhat]

Description

Add metric to log mm embedding calculation time.

Related Issues

Test Plan

Checklist

• I have read the contributing guidelines
• My code follows the project's code style (run bash format.sh)
• I have added tests for my changes (if applicable)
• I have updated the documentation (if applicable)
• My commits include a Signed-off-by: line (DCO compliance)

[github-actions]

👋 Hi! Thank you for contributing to vLLM support on Spyre.
Just a reminder: Make sure that your code passes all the linting checks, otherwise your PR won't be able to be merged. To do so, run ./format.sh.
Now you are good to go 🚀.

We also recommend installing prek and configuring it to check your code before every local commit.

[joerunde]

Do we need to be publishing a non-standard vllm metric here?

[gkumbhat]

I was trying to find a similar metric from vllm that we could reuse, or push the numbers there. Haven't found one yet..

The idea here is to have a way to measure impact of MM processing.

Open to suggestions, if there is a better and more standard way to deal with this.

[gkumbhat]

@joerunde actually usage of create_perf_metric_logger already handles the optional enablement of this metric. So this will only get printed if we pass VLLM_SPYRE_PERF_METRIC_LOGGING_ENABLED env variable.

[joerunde]

Nice!

[joerunde]

lpgtm- from talking with @gkumbhat this is meant for development debugging, not for opsviz on deployed models, so we're using the spyre-specific perf logger here

[joerunde]

ah shoot GK I forgot to hit merge last week, sorry!