docs: changelog and version scoping for 1.1.0 - #163

Release notes and version scoping for codegen 1.1.0, the pair of engine v1.1.0. The release workflow writes only VERSION; this PR carries the rest, so the README (PyPI's long description for 1.1.0) is right at the tag. Rebased on main b3904ca (#164).

What changes

• CHANGELOG.md: a ## 1.1.0 — 2026-10-04 section for the changes since 1.0.1: Emit checked strategy settings and contain C ABI exceptions #159 (emitted C++), Discover the other symbols' feeds a script requests before it runs #164 (request discovery), Render codegen parity README from release-hub facts #160 and docs: keep the anomalies sentence out of a raw HTML block #162 (documentation). 1.1.0 is a minor release: engine v1.1.0 declares six checked settings functions in <pineforge/pineforge.h> (engine Stop an endless counted loop as TradingView does, count float steps by TradingView's measured end rule, accept more keyword and constant forms, and continue a statement after an operator line #317) that the C++ of codegen 1.1.0 defines, and transpile_full() gains requests. Sections: compatibility and migration, checked strategy settings, exceptions at the C ABI, enum input defaults, request discovery, documentation. Discover the other symbols' feeds a script requests before it runs #164's ## Unreleased section is folded into 1.1.0 with its link; no Unreleased heading remains.
• Version scoping, as docs: changelog and version scoping for 1.0.1 #158 did for 1.0.1: README, AGENTS.md and CLAUDE.md (kept twins), CONTRIBUTING.md, docs/PUBLIC_CONTRACT.md and docs/pine-cap-activation.md name 1.1.0 where they named 1.0.1 as the current release or pair; the pairing table has a 1.1.0 row. The "(unreleased)" and "Development builds" wording Discover the other symbols' feeds a script requests before it runs #164 added in README.md and docs/PUBLIC_CONTRACT.md names 1.1.0; the contract's heading is "Request discovery" (the anchor the changelog links), and its result-key table, success envelope example and manifest paragraph list requests and kind.
• Release scoreboard: "Release 1.1.0 grades …" with releases[1.1.0] markers: 7,951 excellent / 38 strong of 7,989 on baseline pineforge-parity-baseline-20261004-engine-7b596622 (engine 7b596622, codegen c5d97ee), rendered from pineforge-release facts/facts.json at eef6d7ec.

Compatibility notes

• Python and JSON contract: additive. transpile_full()'s result and gate/glue.py's transpile_json success envelope gain requests, and input.symbol manifest entries gain "kind": "symbol" (Discover the other symbols' feeds a script requests before it runs #164); no argument, result key or envelope is removed or renamed; transpile() and the error envelope are unchanged. One value changes (Emit checked strategy settings and contain C ABI exceptions #159): for an input.enum whose default is a variable set to a member (defaultSide = Side.long), inputs[].default is "Side.long" where 1.0.1 reported null.
• Report keys (pineforge-app parses them): the engine's docker/run_json.py keeps every key it had with v1.0.1, metrics.equity.sharpe_tv / sortino_tv included. It adds applied_runtime.syminfo = {"qty_step": v, "mincontract": v} when the --syminfo / PINEFORGE_SYMINFO file carries a valid mincontract (engine Keep timestamps 64-bit through function and method parameters, integer array fields and array.from, and in a time input that is changed, copied or kept in a var #322), and applied_runtime.symbol_feeds when --symbol-feeds / PINEFORGE_SYMBOL_FEEDS installs another symbol's bars (engine Codegen: keyword binding, request contexts, array and numeric semantics, drawing collections, diagnostics #327, pineforge-release fix(codegen): escape C++ string literals (quotes/backslash/newline) #23). fingerprint.provenance.runtime equals applied_runtime, so such runs have their own fingerprint; for a run with neither, Keep timestamps 64-bit through function and method parameters, integer array fields and array.from, and in a time input that is changed, copied or kept in a var #322 changes no byte apart from elapsed_seconds. New failure output: one stdout line {"engine":"pineforge","error":...}, harness exit 1, image entrypoint exit 4. Between 1.0.1 and 1.1.0 the fingerprint differs in any case (engine and codegen versions, generated C++ hash), and engine Known issue in PineForge Codegen 1.9.1: str.tostring of built-in boolean flags, and of variables holding them or a ta.* boolean result, returns "1" / "0" #315 / Follow a request's timeframe on every bar, and keep persistent state per timeframe #316 can change a run's trades.

Verified

• lab facts render / lab facts check --repo . --facts https://raw.githubusercontent.com/pineforge-4pass/pineforge-release/eef6d7ecc06c824c2d70e6638c5c36adc23386b5/facts/facts.json (sha256 7c796c3f…): PASS: 15 fact markers; 0 file(s) drifted.
• Discover the other symbols' feeds a script requests before it runs #164's C++ note, re-checked on this branch: the engine corpus's 325 sources and the 540 test-fixture .pine files, one fresh process per file (PYTHONHASHSEED=0), c5d97ee against b3904ca: 0 differ, 51 refused with the same message.
• Discover the other symbols' feeds a script requests before it runs #164's README and contract examples match transpile_full() and transpile_json output for tests/fixtures/request_discovery/case_strategy.pine (the line-7 entry, same fields and order); the error envelope carries no requests.
• Earlier on this branch, for Emit checked strategy settings and contain C ABI exceptions #159: the 325 corpus sources and 277 gate fixtures transpile, with the settings code removed (tests/_legacy_cpp.py), to 1.0.1's C++ byte for byte; the README tutorial was re-run on a Linux x86_64 build box (9 trades, +738.20; 13 trades with the old sizing declared); timing on CPython 3.14.6, Apple M4 Max.
• git diff --check; twine check --strict on a fresh sdist and wheel.

🤖 Generated with Claude Code