feat(agent): implement model tier selection with variant support for subagents

[mcking-07]

What does this PR do?

Closes #6651

The issue is that subagents called via Task tool can't dynamically select models based on task complexity. Users end up creating duplicate agents (explore-quick, explore-standard, etc.) just to vary models, or waste money using powerful models for simple tasks.

I implemented the unified approach (Solution C) from the issue: added model_tier parameter to Task tool with quick/standard/advanced values. You can configure tier-to-model mappings globally or per-agent, with agent-level overriding global when defined.

Resolution hierarchy:

1. Agent's explicit model (if set, tier ignored)
2. Agent-level model_tiers[tier] - if defined
3. Global model_tiers[tier] - if defined
4. Parent session model (fallback)

I also added optional variant support to tier configurations (as noted in a later comment on the issue), allowing models to use different reasoning modes.

See #6651 for detailed implementation discussion and configuration examples.

How did you verify your code works?

Tests: Added comprehensive tests covering model tier resolution and variant extraction - 720 total pass, with 12 new tests for the feature (agent tier priority, global tier fallback, variant parsing, and resolution hierarchy).

Manual testing: Have been using this setup locally for about 2 weeks for development work. Configured opencode.json with model_tiers for build and plan agents with different variants. Called Task with different model_tier values daily and verified correct models were used. Confirmed agent tiers override global tiers as expected.

[github-actions]

The following comment was made by an LLM, it may be inaccurate:

Potentially Related PRs Found:

1. Add default variant config per agent (Add default variant config per agent #7140)

   • Add default variant config per agent #7140
   • Relation: Addresses variant configuration for agents, which is related to the variant support being added in this PR.

2. feat: add variant support for subagents (feat: add variant support for subagents (#7138) [alt of #7140] #7156)

   • feat: add variant support for subagents (#7138) [alt of #7140] #7156
   • Relation: Previous work on variant support for subagents; this PR builds upon variant functionality.

3. feat: add variant support for commands (feat: add variant support for commands (#7713) #7907)

   • feat: add variant support for commands (#7713) #7907
   • Relation: Related variant support implementation for commands; may have similar patterns to consider.

4. feat(task): Add subagent-to-subagent delegation with budgets, persistent sessions, and hierarchical session navigation (feat(task): Add subagent-to-subagent delegation with budgets, persistent sessions, and hierarchical session navigation #7756)

   • feat(task): Add subagent-to-subagent delegation with budgets, persistent sessions, and hierarchical session navigation #7756
   • Relation: Earlier work on task delegation and subagent handling; may interact with the model tier selection feature.

These are historical variants/subagent work rather than current duplicates. The current PR (11377) appears to be the primary implementation for model tier selection.

[RicardoMonteiroSimoes]

If I understand this correctly, it just allows us to define levels for the same model, and allow references of it in agent definitions, right?
Something like a hierarchical selection of (depending on whats available) models isn't possible, as a simple example:

models: [opus, sonnet, gpt-5.3]

Which would enable per project provider locks but still allow transferability of agents across projects?