Addresses #5627. Skip clean runs during output flush

[YourRobotOverlord]

Summary

Optimize OutputBase.Write so sparse dirty rows skip clean gaps instead of emitting one cursor-position operation per clean cell.

• Closes OutputBase.Write issues one cursor-move console write per clean cell, making sparse frames far more expensive than full ones #5627
• Related work: Optimize output flushing by synchronizing dirty-row flags #5601 and Fixes #5601. Synchronize dirty-row flags after output #5602 skip wholly clean rows; this PR handles clean gaps within dirty rows

Changes

• Accumulate and flush consecutive dirty cells as output runs
• Position the cursor once at the beginning of each dirty run
• Preserve hyperlink, raster, wide-grapheme, legacy-console, and dirty-row behavior
• Add 270x72 full-frame and sparse-border BenchmarkDotNet coverage
• Add a Windows ConPTY benchmark command and regression tests

Benchmarks

Case	Before	After	Result
Sparse BenchmarkDotNet mean	2.741 ms	0.704 ms	3.89x faster
Sparse allocation	3.83 MiB	300.53 KiB	92.3% lower
Sparse cursor calls	19,440	144	135x fewer
ConPTY median	347.196 ms	5.713 ms	60.8x faster
ConPTY p95	353.920 ms	6.070 ms	58.3x faster

Full-frame BenchmarkDotNet performance remained effectively flat. Content writes remain unchanged at 144 for the sparse frame.

Testing

• Debug build: 0 warnings, 0 errors
• OutputBaseTests: 73 passed
• Parallelizable unit tests: 17,575 passed, 17 skipped
• Non-parallelizable unit tests: 72 passed, 2 skipped
• Integration tests: 437 passed
• ConPTY captures used tuirec v0.9.1 at 270x72

To pull down this PR locally:

git remote add yor https://github.com/YourRobotOverlord/Terminal.Gui.git
git fetch yor fix/skip-clean-runs
git checkout --track yor/fix/skip-clean-runs

[YourRobotOverlord]

Before:

After:

[Copilot]

Pull request overview

This PR improves OutputBase.Write performance for sparse frames by skipping clean gaps within dirty rows, reducing cursor-position operations and console writes while preserving existing hyperlink/raster/wide-grapheme behaviors. It also adds benchmarks and regression tests to validate the new flush strategy and quantify performance changes.

Changes:

• Update OutputBase.Write to flush consecutive dirty-cell runs and skip clean gaps without per-cell cursor moves.
• Add unit tests validating cursor-move and dirty-flag behavior for sparse and multi-run output.
• Add benchmark coverage (BenchmarkDotNet + ConPTY mode) and documentation for measuring output flush performance.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
Terminal.Gui/Drivers/Output/OutputBase.cs	Implements skipping clean gaps within dirty rows and run-based cursor positioning/flush logic.
Tests/UnitTestsParallelizable/Drivers/Output/OutputBaseTests.cs	Adds regression tests for sparse-border and multi-run dirty output; extends test output instrumentation.
Tests/Benchmarks/ConsoleDrivers/OutputBuffer/OutputWriteBenchmark.cs	Introduces benchmarks for full-frame vs sparse-frame flush, plus a Windows ConPTY timing mode.
Tests/Benchmarks/Program.cs	Adds a CLI entry point for running the ConPTY output benchmark (output-conpty).
Tests/Benchmarks/README.md	Documents how to run the new output flush benchmarks, including ConPTY instructions.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[tig]

Well done. Thank you!