[GSoC] Update `tune` API for LLM hyperparameters optimization #782

Job	Run time
e2e (v1.27.11, long-running-resume,from-volume-resume,median-stop)	5m 48s
e2e (v1.27.11, grid,bayesian-optimization,tpe,multivariate-tpe,cma-es,hyperband)	5m 46s
e2e (v1.27.11, file-metrics-collector,pytorchjob-mnist)	5m 46s
e2e (v1.27.11, median-stop-with-json-format,file-metrics-collector-with-json-format)	5m 33s
e2e (v1.28.7, long-running-resume,from-volume-resume,median-stop)	5m 47s
e2e (v1.28.7, grid,bayesian-optimization,tpe,multivariate-tpe,cma-es,hyperband)	5m 48s
e2e (v1.28.7, file-metrics-collector,pytorchjob-mnist)	5m 46s
e2e (v1.28.7, median-stop-with-json-format,file-metrics-collector-with-json-format)	5m 46s
e2e (v1.29.2, long-running-resume,from-volume-resume,median-stop)	5m 45s
e2e (v1.29.2, grid,bayesian-optimization,tpe,multivariate-tpe,cma-es,hyperband)	5m 46s
e2e (v1.29.2, file-metrics-collector,pytorchjob-mnist)	5m 47s
e2e (v1.29.2, median-stop-with-json-format,file-metrics-collector-with-json-format)	5m 48s
	1h 9m 6s

Provide feedback