Reduce flash size of TetrisAI_V2 by 97%

[gustebeast]

Main branch without Tetris
Flash: [======== ] 79.8% (used 1255301 bytes from 1572864 bytes)

Main branch with Tetris (+196kb)
Flash: [========= ] 92.3% (used 1452049 bytes from 1572864 bytes)

This commit with Tetris (+6kb, 97% less flash)
Flash: [======== ] 80.2% (used 1261625 bytes from 1572864 bytes)

Summary by CodeRabbit

Release Notes

• Refactor

  • Optimized internal code structure in Tetris AI module by removing unused dependencies and streamlining queue management implementation.

• Chores

  • Removed a public method from TetrisBag class; verify compatibility if your code relies on queue piece indexing functionality.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

Walkthrough

This PR removes unused include dependencies from three header files in the TetrisAI_v2 module and refactors the TetrisBag class by removing a public method overload and replacing std::rotate with a manual loop implementation.

Changes

Cohort / File(s)	Summary
Dependency cleanup usermods/TetrisAI_v2/gridbw.h, usermods/TetrisAI_v2/pieces.h	Removed unused include directives (<iterator> and <iostream> respectively); no logic or API changes.
API and implementation refactoring usermods/TetrisAI_v2/tetrisbag.h	Removed public method overload void queuePiece(uint8_t idx), refactored void queuePiece() to use manual loop instead of std::rotate, and removed <algorithm> include.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately describes the main objective of the PR—reducing flash usage of the TetrisAI_V2 module through dependency removal and code optimization.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

• 📝 Generate docstrings

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@usermods/TetrisAI_v2/tetrisbag.h`:
- Around line 86-92: queuePiece() can underflow when the queue is empty because
piecesQueue.size() - 1 will wrap and cause out-of-range access; fix by returning
early (or asserting) when the queue is empty (e.g., check piecesQueue.empty() or
queueLength == 0) before shifting elements and assigning getNextPiece(),
ensuring you guard the piecesQueue access in queuePiece().

[coderabbitai]

⚠️ Potential issue | 🟡 Minor

Guard against empty queue to avoid out-of-range access.

If queueLength == 0, piecesQueue.size() - 1 underflows and the assignment is out-of-bounds. Add a quick empty check (or assert) to keep this safe.

🐛 Suggested fix

 void queuePiece()
 {
+    if (piecesQueue.empty()) return;
     //move vector to left
     for (uint8_t i = 1; i < piecesQueue.size(); i++) {
         piecesQueue[i - 1] = piecesQueue[i];
     }
     piecesQueue[piecesQueue.size() - 1] = getNextPiece();
 }

🤖 Prompt for AI Agents

In `@usermods/TetrisAI_v2/tetrisbag.h` around lines 86 - 92, queuePiece() can
underflow when the queue is empty because piecesQueue.size() - 1 will wrap and
cause out-of-range access; fix by returning early (or asserting) when the queue
is empty (e.g., check piecesQueue.empty() or queueLength == 0) before shifting
elements and assigning getNextPiece(), ensuring you guard the piecesQueue access
in queuePiece().

[Aircoookie]

Disregard, for loop is never entered if piecesQueue.size() < 1

[coderabbitai]

@Aircoookie You're absolutely right that the for loop won't execute when the queue is empty. I was also concerned about the assignment on the next line (piecesQueue[piecesQueue.size() - 1] = getNextPiece();), which could underflow if size() == 0. However, if the queue is always initialized with a non-zero size in practice, that wouldn't be an issue. I'll defer to your judgment on this! 👍

[Aircoookie]

LGTM, thank you! Awesome that we can achieve such huge flash savings by avoiding these imports