feat: add rotating goal data model

[itsmiso-ai]

Fixes #142

Add a minimal pure data model for rotating colony goals. No UI yet.

Goal state structure

• goal id/type — unique identifier and type enum (resource/build/build_complete)
• target resource/build/task — target specification per goal type
• target amount — threshold for completion
• current progress — computed or updated value
• completion state — boolean flag set by complete_goal()

Goal catalog (7 fixed entries)

ID	Type	Target
gather_wood	resource	10 wood
gather_stone	resource	5 stone
gather_food	resource	8 food
build_hut	build	hut
build_workshop	build	workshop
build_garden	build	garden
any_build	build_complete	any build

Acceptance criteria

• Goal catalog exists in a testable place (scripts/rotating_goal.gd)
• Active goal can be selected deterministically for tests (select_next_active_goal)
• Progress can be updated or computed from existing game state
• Completion detection is covered by tests
• No UI changes yet
• Link back to design: add compact rotating colony goals #131

Refs: #131

[its-miso]

✅ Automated recommendation: APPROVE

Analysis engine: MiniMax-M2.7@https://litellm.jory.dev/v1 (anthropic)

PR Review: feat: add rotating goal data model

Summary

This PR adds a pure data model for rotating colony goals in scripts/rotating_goal.gd with corresponding tests in tests/test_rotating_goal.gd. The implementation is focused, well-documented, and includes comprehensive test coverage.

Change-by-Change Findings

scripts/rotating_goal.gd (120 lines, ADDED)

• Goal types: Three constants define GOAL_TYPE_RESOURCE, GOAL_TYPE_BUILD, and GOAL_TYPE_BUILD_COMPLETE
• Goal catalog: Seven fixed entries covering all specified concepts (wood, stone, food, hut, workshop, garden, any_build)
• apply_goal_template: Correctly uses duplicate(true) for deep copying targets, preventing catalog mutation
• select_next_active_goal: Returns first non-completed goal deterministically, or empty dict when all complete
• Progress helpers: Separate functions for updating (update_resource_progress) and computing (compute_*_progress) progress from game state
• Completion detection: Type-dispatching is_goal_complete() with individual is_*_complete() checks
• Reward behavior: complete_goal() sets completed=true with no reward field (matches issue requirement)

tests/test_rotating_goal.gd (234 lines, ADDED)

• Catalog tests: Verifies 7 entries, all goal types, all goal IDs
• Template tests: Validates deep copy behavior (template_deep_copies_target)
• Selection tests: Covers first goal, skipping completed, last remaining, all-complete edge case
• Progress tests: Validates clamping at target amount, game state computation, missing resource handling
• Completion tests: Tests all three goal types with progress=0 and progress>=threshold cases
• No-op reward test: Explicitly verifies no reward field is added on completion

Standards Compliance

✅ Repository conventions followed:

• GDScript class with class_name RotatingGoal (matches layout_math.gd pattern)
• Constants use SCREAMING_SNAKE_CASE
• Static utility functions for pure data operations
• Inline comments with ## style documentation
• Deep-copy defensive programming for catalog data

✅ Issue #142 requirements fully met:

• Goal state structure includes id/type, target, target amount, current progress, completion state
• Goal catalog uses existing game concepts only (wood, stone, food, hut, workshop, garden)
• No UI changes, no rewards beyond no-op completion
• References design: add compact rotating colony goals #131 as requested

Linked Issue Fit

✅ Issue #142 acceptance criteria status:

• Goal catalog exists in a testable place → scripts/rotating_goal.gd
• Active goal can be selected deterministically → select_next_active_goal(completed_ids)
• Progress can be updated or computed from game state → update_resource_progress, compute_*_progress
• Completion detection covered by tests → test_is_goal_complete with 7 assertions
• No UI changes yet → Pure data model only
• Link back to design: add compact rotating colony goals #131 → Commented at line 3

Evidence Provider Findings

No evidence providers configured.

Tool Harness Findings

No tool requests executed (planned: 0, executed: 0).

Unknowns / Needs Verification

None. Implementation is complete and self-contained.