Fail loudly on canonical continuous future orders, add Future.Canonical and mapped-contract guidance

[jhonabreul]

Description

Ordering the canonical continuous futures symbol fails silently: CalculateOrderQuantity(canonical, 1.0) returns a plausible quantity, and SetHoldings(canonical, 0.5) submits an order rejected with only:

The security with symbol '/ES' is marked as non-tradable.

Related traps: Future.Mapped is null during Initialize, so passing it to order methods dies with a bare NRE; stop/limit prices computed from adjusted continuous prices sit far from the mapped contract's raw prices.

Cause: nothing on the canonical order path names the problem or points to Future.Mapped, and the buying power path computes quantities for a security that can never trade.

The fix:

• PortfolioTarget.Percent rejects canonical future symbols with an instructive error and returns null, so CalculateOrderQuantity returns 0 and SetHoldings no longer submits doomed orders. Restricted to futures.
• QCAlgorithm.PreOrderChecksImpl non-tradable rejection now uses Messages.QCAlgorithm.CanonicalSymbolNotTradable for canonical symbols, telling the user to trade Future.Mapped. Messages are language-aware for Python.
• QCAlgorithm.GetSecurityForOrder throws ArgumentNullException with the same guidance for null symbols instead of an opaque NRE. Covers all order helpers.
• Future.Canonical added as an alias of Symbol.Canonical (future.canonical in Python); Future.Mapped XML docs now state it is null until the first universe selection.
• PreOrderChecksImpl warns once when a future order's stop/limit/trigger price deviates more than 10% from market price while the continuous subscription uses a non-Raw DataNormalizationMode.
• SignalExportTargetTests future cases switched from holdings on the canonical security (now rejected) to a real ES contract.

Not changed: slice.Bars lookups with the canonical symbol already resolve the continuous bar (asserted in the regression algorithm).

Related Issue

N/A

Note: sibling PR #9675 covers the null-symbol NRE in AddConsolidator/GetSubscription; this PR covers the order path.

Motivation and Context

The silent canonical-order failure is especially costly to debug. Loud, named errors close the worst silent failure mode in futures algorithms.

Requires Documentation Change

No — error messages and a new property alias.

How Has This Been Tested?

• New regression algorithm ContinuousFutureCanonicalOrdersRegressionAlgorithm (C# + Python): asserts the new errors and null/zero results end-to-end, that canonical bars still resolve, and that the mapped contract trades normally.
• New AlgorithmTradingTests covering canonical order rejection, zero order quantity, null-symbol guidance, and the price-deviation warning (fires once for non-Raw, never for Raw or near-market prices).
• PortfolioTargetTests.PercentReturnsNullForCanonicalFutureSymbol.
• Suites: AlgorithmTradingTests + all Framework.Portfolio tests: 813 passed, 0 failed. ContinuousFuture*/FuturesFramework* regression algorithms: all passed.

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)

Breaking only narrowly: PortfolioTarget.Percent/CalculateOrderQuantity now return null/0 with an error for canonical future symbols instead of a quantity that could not be traded anyway.

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>