fix(run_agent): read context_length from providers/custom_providers for compression model

[ms-alan]

Summary

When auxiliary.compression.model uses a custom provider with per-model context_length configured under providers/<key>/models/<model>/context_length, the compression feasibility check now looks up that value — mirroring the main model's existing providers lookup in __init__.

Root cause: _aux_compression_context_length_config was only populated from auxiliary.compression.context_length (explicit override). The same providers / custom_providers per-model context_length lookup that exists for the main model was missing for the auxiliary model.

Fix: In _prepare_compression(), added providers/custom_providers context_length lookup for the auxiliary compression model (mirrors the main model lookup in init`)

Testing

1. Configure auxiliary.compression.provider: custom + auxiliary.compression.model pointing to a local endpoint with context_length: 262141 in providers.local-kimi.models.kimi-code/kimi-code.context_length
2. Start a session — the compression threshold should use 262141 instead of the endpoint's auto-detected 128000

Closes #13807

[alt-glitch]

Related to #13540 which also targets the same compression context_length propagation gap via a different approach.

[alt-glitch]

Related to #13540 which also targets the same compression context_length propagation gap via a different approach.

[smallerxie]

Related: PR #17460 addresses the same root cause but with a more general approach. Instead of adding a providers lookup in _prepare_compression() (compression-only), it auto-loads custom_providers inside get_model_context_length() — fixing all auxiliary paths at once.

See also: #13540, #12977, #13807

[teknium1]

Thanks for identifying the auxiliary compression context-length propagation gap. This is now implemented on current main; this is an automated hermes-sweeper review.

• Commit 7becb19ea00c13bdff6f78b71aa3ddfb0bdb5378 forwards the agent's compatible custom-provider configuration to compression context-length detection.
• agent/conversation_compression.py:236-245 now passes custom_providers=agent._custom_providers to get_model_context_length() for the actual auxiliary compression client.
• agent/agent_init.py:1653-1665 builds that list through get_compatible_custom_providers(), whose compatibility layer covers both legacy custom_providers and keyed providers configuration.

The later shared-path fix satisfies the PR's requested behavior without duplicating provider lookup logic in the compression check.