feat(config): add longcontx profile to codex config.toml

[konard]

Summary

• Moves model_context_window = 1050000 and model_auto_compact_token_limit = 945000 from the global config.toml section into a new [profiles.longcontx] profile
• The global section now only sets model, reasoning effort, and policies — no forced context-window overrides for every model
• Users can opt into large-context mode via codex --profile longcontx

Changes

packages/lib/src/core/templates-entrypoint/codex.ts
Updated the shell template that writes ~/.codex/config.toml on container startup.

packages/lib/src/usecases/auth-sync-helpers.ts
Updated defaultCodexConfig (used by auth-sync to create/rewrite the config) to match the new layout.

Mathematical Guarantees

Invariants:

• ∀c: default_config(c) → longcontx_profile(c) ∧ ¬global_context_override(c)
• Context-window overrides are profile-scoped, not global

Resulting config.toml structure:

# docker-git codex config
model = "gpt-5.4"
model_reasoning_effort = "xhigh"
plan_mode_reasoning_effort = "xhigh"
personality = "pragmatic"

approval_policy = "never"
sandbox_mode = "danger-full-access"
web_search = "live"

[features]
shell_snapshot = true
multi_agent = true
apps = true
shell_tool = true

[profiles.longcontx]
model = "gpt-5.4"
model_context_window = 1050000
model_auto_compact_token_limit = 945000
model_reasoning_effort = "xhigh"
plan_mode_reasoning_effort = "xhigh"

Test plan

• All 186 tests pass locally (pnpm --filter ./packages/lib test + pnpm --filter ./packages/app test)
• Existing tests for model_context_window and model_auto_compact_token_limit presence still pass (values remain in profile)
• No changes to test files required

Fixes #183

🤖 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: $0.976229
• Calculated by Anthropic: $0.920835 USD
• Difference: $-0.055394 (-5.67%)

🤖 Models used:

• Tool: Claude
• Requested: sonnet
• Model: Claude Sonnet 4.6 (claude-sonnet-4-6)

📎 Log file uploaded as Gist (819KB)

• 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:

?? issue-183.txt
?? pr-184.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: $0.293488
• Calculated by Anthropic: $0.127048 USD
• Difference: $-0.166440 (-56.71%)

🤖 Models used:

• Tool: Claude
• Requested: sonnet
• Model: Claude Sonnet 4.6 (claude-sonnet-4-6)

📎 Log file uploaded as Gist (968KB)

• 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 triggered (iteration 1)

Reason: CI failures detected

Starting new session to address the issues.

---

Auto-restart-until-mergeable mode is active. Will continue until PR becomes mergeable.

[konard]

🔄 Auto-restart-until-mergeable Log (iteration 1)

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

💰 Cost estimation:

• Public pricing estimate: $0.279248
• Calculated by Anthropic: $0.616813 USD
• Difference: $0.337565 (+120.88%)

🤖 Models used:

• Tool: Claude
• Requested: sonnet
• Model: Claude Sonnet 4.6 (claude-sonnet-4-6)

📎 Log file uploaded as Gist (1490KB)

• 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 triggered (iteration 2)

Reason: Uncommitted changes detected

Starting new session to address the issues.

---

Auto-restart-until-mergeable mode is active. Will continue until PR becomes mergeable.

[konard]

🔄 Auto-restart-until-mergeable Log (iteration 2)

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

💰 Cost estimation:

• Public pricing estimate: $0.428841
• Calculated by Anthropic: $0.240109 USD
• Difference: $-0.188731 (-44.01%)

🤖 Models used:

• Tool: Claude
• Requested: sonnet
• Model: Claude Sonnet 4.6 (claude-sonnet-4-6)

📎 Log file uploaded as Gist (1753KB)

• 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