feat: replace deprecated qwen3.6-plus-free default with nemotron-3-super-free (#242)

[konard]

Summary

OpenCode Zen ended the free promotion for qwen3.6-plus-free (Qwen 3.6 Plus Free) in April 2026. The model now returns HTTP 401 with "Free promotion has ended for Qwen3.6 Plus Free", breaking the default agent experience for all free-tier users.

• Default model changed: opencode/qwen3.6-plus-free → opencode/nemotron-3-super-free (~262K context, NVIDIA hybrid Mamba-Transformer)
• Compaction cascade updated: removed deprecated qwen3.6-plus-free, reordered remaining models by context size
• Documentation updated: moved qwen3.6-plus-free to deprecated/discontinued sections in FREE_MODELS.md, MODELS.md, README.md
• Case study added: docs/case-studies/issue-242/ with full root cause analysis and evidence from failed run logs
• Changeset added: minor version bump via changesets (will be applied automatically by release pipeline)

Files Changed

File	Change
js/src/cli/defaults.ts	New DEFAULT_MODEL, updated compaction cascade
js/src/cli/argv.ts	Updated compaction models comment
js/src/provider/provider.ts	Updated priority lists (removed qwen3.6-plus-free)
js/.changeset/replace-default-model-242.md	Changeset for automated release
FREE_MODELS.md	Moved qwen3.6-plus-free to discontinued
MODELS.md	Updated default model, added to discontinued
README.md	Updated model references
js/tests/compaction-model.test.ts	Updated default model assertions
js/tests/model-strict-validation.test.ts	Updated default model reference
docs/case-studies/issue-242/	Case study with log analysis

Root Cause

OpenCode Zen API now returns:

{"type":"error","error":{"type":"ModelError","message":"Free promotion has ended for Qwen3.6 Plus Free. You can continue using the model by subscribing to OpenCode Go - https://opencode.ai/go"}}

Test plan

• Unit tests pass (compaction-model and model-strict-validation test files)
• Changeset validation passes locally
• Pre-commit hooks pass (prettier formatting)
• CI pipeline passes

Fixes #242

🤖 Generated with Claude Code

[konard]

🤖 Solution Draft Log

This log file contains the complete execution trace of the AI solution draft process.

💰 Cost estimation:

• Public pricing estimate: $3.286674
• Calculated by Anthropic: $3.286674 USD
• Difference: $0.000000 (+0.00%)

📊 Context and tokens usage:

• Context window: 99.9K / 1M (10%) input tokens, 19.6K / 128K (15%) output tokens

Total: (95.4K + 4.4M cached) input tokens, 19.6K output tokens, $3.286674 cost

🤖 Models used:

• Tool: Anthropic Claude Code
• Requested: opus
• Model: Claude Opus 4.6 (claude-opus-4-6)

📎 Log file uploaded as Gist (1609KB)

• View complete solution draft log

---

Now working session is ended, feel free to review and add any feedback on the solution draft.

[konard]

🔄 Auto-restart 1/3

Detected uncommitted changes from previous run. Starting new session to review and commit or discard them.

Uncommitted files:

?? previous-run-log.txt

---

Auto-restart will stop after changes are committed or discarded, or after 2 more iterations. Please wait until working session will end and give your feedback.

[konard]

🔄 Auto-restart 1/3 Log

This log file contains the complete execution trace of the AI solution draft process.

💰 Cost estimation:

• Public pricing estimate: $0.806472
• Calculated by Anthropic: $0.806472 USD
• Difference: $0.000000 (0.00%)

📊 Context and tokens usage:

• Context window: 41.4K / 1M (4%) input tokens, 5.6K / 128K (4%) output tokens

Total: (30.5K + 954.1K cached) input tokens, 5.6K output tokens, $0.806472 cost

🤖 Models used:

• Tool: Anthropic Claude Code
• Requested: opus
• Model: Claude Opus 4.6 (claude-opus-4-6)

📎 Log file uploaded as Gist (3106KB)

• View complete solution draft log

---

Now working session is ended, feel free to review and add any feedback on the solution draft.

[konard]

✅ Ready to merge

This pull request is now ready to be merged:

• All CI checks have passed
• No merge conflicts
• No pending changes

---

Monitored by hive-mind with --auto-restart-until-mergeable flag