Polish the order surface: numeric OrderFee, flat combo group ids, tag tolerance and notional targeting

[jhonabreul]

Description

Python algorithms repeatedly crash on the first names they reach for on the order surface. For example:

'OrderFee' object has no attribute 'amount'
Trying to perform a summation ... between 'OrderFee' and 'OrderFee' objects throws a TypeError
'ComboLimitOrder' object has no attribute 'group_order_manager_id'
self.liquidate("EOD close") -> overload dump (the tag binds to the symbol slot)

The change adds the flat shortcuts and operators these calls expect:

• OrderFee.Amount / Currency shortcuts, plus arithmetic and comparison operators delegating to Value.Amount, so fee1 + fee2, sum(fees) and fee > 0 work from Python.
• OrderEvent.OrderFeeAmount, OrderEvent.GroupId and Order.GroupOrderManagerId shortcuts.
• ComboOrderTicket returned by the combo order methods: holds the leg tickets and aggregates the group (Tickets, GroupOrderManagerId, Filled). It subclasses List<OrderTicket>, so existing iteration/indexing keeps working in both languages.
• market_order(symbol, 10, "entry"): Python-only overload accepting the tag in the third positional slot.
• Liquidate(string): resolves known tickers; unknown strings fail with a prescriptive error (pointing at Liquidate(tag: ...)) instead of a silent no-op or an overload dump.
• OrderTargetNotional(symbol, targetNotional): market order adjusting holdings to an absolute notional; handles contract multipliers, lot-size rounding, and returns null when already at target.

Deferred: an OrderStatus.REJECTED alias for INVALID (duplicate enum values change either the log or the JSON surface), and float(order_fee) (needs an interop-layer __float__ slot; the Amount shortcut covers it).

Related Issue

N/A

Motivation and Context

These are the exact names and operations users try first, and each was a recurring source of failed backtests.

Requires Documentation Change

The new public members are candidates for documentation once merged.

How Has This Been Tested?

• New OrderSurfaceShortcutsRegressionAlgorithm (C# + Python): asserts every shortcut, the combo ticket aggregation, the positional tag, the Liquidate error and OrderTargetNotional. Identical statistics in both languages.
• New unit tests: OrderFeeTests (shortcuts, operators, JSON shape unchanged), ComboOrderTicketTests, OrderEventTests/OrderTests shortcut cases, AlgorithmTradingTests (Liquidate and OrderTargetNotional math incl. futures multiplier).
• A scratch Python probe reproduced all the error signatures on master and confirmed them fixed after the change.
• Suites: Tests.Common.Orders 994, Tests.Algorithm 12,719, Tests.Python 1,297, combo/liquidate regression algorithms 24 — all passed.

Compatibility note: the combo order methods now return ComboOrderTicket (a List<OrderTicket> subclass — source compatible), and Liquidate with an unknown ticker now raises instead of warning-and-ignoring.

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• Refactor (non-breaking change which improves implementation)
• Performance (non-breaking change which improves performance. Please add associated performance test and results)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Non-functional change (xml comments/documentation/etc)

Checklist:

• My code follows the code style of this project.
• I have read the CONTRIBUTING document.
• I have added tests to cover my changes.
• All new and existing tests passed.
• My branch follows the naming convention bug-<issue#>-<description> or feature-<issue#>-<description>