[SPARK-50194][SS][PYTHON] Integration of New Timer API and Initial State API with Timer

[jingz-db]

What changes were proposed in this pull request?

As Scala side, we modify the timer API with a separate handleExpiredTimer function inside StatefulProcessor, this PR make a change to the timer API to couple with API on Scala side. Also adds a timer parameter to pass into handleInitialState function to support use cases for registering timers in the first batch for initial state rows.

Why are the changes needed?

This change is to couple with Scala side of APIs: #48553

Does this PR introduce any user-facing change?

Yes.
We add a new user defined function to explicitly handle expired timeres:

def handleExpiredTimer(
        self, key: Any, timer_values: TimerValues, expired_timer_info: ExpiredTimerInfo

We also add a new timer parameter to enable users to register timers for keys exist in the initial state:

def handleInitialState(
        self,
        key: Any,
        initialState: "PandasDataFrameLike",
        timer_values: TimerValues) -> None

How was this patch tested?

Add a new test in test_pandas_transform_with_state

Was this patch authored or co-authored using generative AI tooling?

No