Custom FIll Model not trigger for stopmarket order

Reviewing Jesse's Python code for bugs, here are the identified issues:

1. Custom Fill Model Implementation: The custom fill model MyFillModel is correctly defined but not correctly set for the option contract. The method SetFillModel is used on self.activeOption, which is returned by AddOptionContract. However, AddOptionContract returns a Security object, not an Option object. The SetFillModel method should be called on the Option object itself, not on the returned value from AddOptionContract. This might be why the custom fill model seems not to be used.

2. Time Check in OnData: The time check if self.Time.hour == 16 and self.Time.minute == 0: is used to execute code at a specific time. However, this approach might not be reliable due to the possibility of data arriving slightly before or after the exact minute. It's better to use scheduled events for precise timing.

3. Stop Market Order Quantity: In the line self.stopMarketOrderTicket = self.StopMarketOrder(self.activeOption.Symbol, 1, 29, tag="StopMarket Order"), the quantity is set to 1 for the stop market order. If the intention is to close the position opened by the market order, the quantity should match but with the opposite sign. If the market order is for selling (-1), the stop market order should be for buying (1), which is correct in this context. However, if the position size changes, this needs to be adjusted accordingly.

4. Lack of Order Event Handling Logic: In the OnOrderEvent method, there's no implementation provided. This method should include logic to handle different order events, such as logging or adjusting strategy based on order fills, partial fills, cancellations, etc.

5. Use of datetime.min: While not directly a bug in the provided code, using datetime.min as a placeholder for self.entryTime and self.exitTime might lead to confusion or errors in date comparisons later in the algorithm. It's usually better to use None for uninitialized dates/times and check for None before performing date/time operations.

6. Error in Research Notebook: In the research notebook (research.ipynb), there's an inconsistency in the dates used for start_time and end_time variables. The end_time is set to a date before the start_time, which would result in an error or empty DataFrame when querying historical data. This is not directly related to the algorithm but could affect the analysis or testing of the strategy.

7. Missing Logic for Active Option Management: The code snippet does not show how the algorithm manages or tracks the active option (self.activeOption) beyond placing the initial market and stop market orders. For a complete strategy, there should be logic to handle the expiration of the option, any assignment or exercise events, and possibly selecting a new option to trade as the strategy progresses.

These points highlight areas in the code that may need attention or correction to ensure the algorithm functions as intended, especially regarding the custom fill model's application and the precise timing of actions within the OnData method.

Calling SetFillModel on the option object itself produced the same results.  The custom fill model is still not being used.

After reviewing Jesse's updated Python code and considering the issue with the custom fill model still not being used, here are some potential issues and considerations:

1. Custom Fill Model Method Signature: The custom fill model MyFillModel correctly overrides the StopMarketFill method. However, it's important to ensure that the method signature exactly matches what the LEAN engine expects. The method signature seems correct, but it's crucial to verify against the latest documentation or source code, as discrepancies can lead to the method not being called.

2. Setting the Fill Model on the Correct Security Object: Jesse mentioned that calling SetFillModel on the option object itself produced the same results. It's important to ensure that the fill model is set on the correct security object. In the provided code, the fill model is set on the option chain object (self.option) before any contracts are added. This is correct, but it's also necessary to ensure that the fill model is applied to the specific option contract that is being traded. In the OnData method, after selecting the specific option contract (self.activeOption = self.AddOptionContract(option.Symbol)), it might be necessary to explicitly set the fill model again for this specific contract if the LEAN engine does not automatically inherit the fill model from the option chain.

3. Verification of Order Types: Ensure that the custom fill model's StopMarketFill method is indeed applicable to the types of orders being placed. The method name suggests it is specifically for stop market orders. If the orders being placed are of a different type, or if there's any mismatch in how orders are being identified and processed, the custom fill model might not be invoked as expected.

4. Debugging and Logging: To diagnose why the custom fill model is not being used, add logging statements within the StopMarketFill method to confirm whether it's being called. This can help identify if the issue is with the method not being invoked or if the logic within the method is not executing as expected.

5. Review LEAN Engine Updates: If the issue persists despite the above considerations, review any recent updates or changes to the LEAN engine that might affect how custom fill models are applied or invoked. The QuantConnect forum and GitHub repository might have relevant discussions or documentation on this topic.

6. Consult QuantConnect Support: If all else fails and the custom fill model still does not seem to be used, consider reaching out to QuantConnect support or the community forum with detailed information about the issue. The QuantConnect community and support team can provide specific advice based on the latest platform capabilities and best practices.

Given the information provided, these are general suggestions to troubleshoot the issue. Without access to the full context of how the LEAN engine is currently handling custom fill models, these are starting points for further investigation.