fix(context): honor custom_providers context_length on /model switch + bump probe tier to 256K (#15844)

Fixes #15779. Custom-provider per-model context_length (`custom_providers[].models.<id>.context_length`) is now honored across every resolution path, not just agent startup. Also adds 256K as the top probe tier and default fallback.

## What changed

New helper `hermes_cli.config.get_custom_provider_context_length()` — single source of truth for the per-model override lookup, with trailing-slash-insensitive base-url matching.

`agent.model_metadata.get_model_context_length()` gains an optional `custom_providers=` kwarg (step 0b — runs after explicit `config_context_length` but before every other probe).

Wired through five call sites that previously either duplicated the lookup or ignored it entirely:
- `run_agent.py` startup — refactored to use the new helper (dedups legacy inline loop, keeps invalid-value warning)
- `AIAgent.switch_model()` — re-reads custom_providers from live config on every /model switch
- `hermes_cli.model_switch.resolve_display_context_length()` — new `custom_providers=` kwarg
- `gateway/run.py` /model confirmation (picker callback + text path)
- `gateway/run.py` `_format_session_info` (/info)

## Context probe tiers

`CONTEXT_PROBE_TIERS = [256_000, 128_000, 64_000, 32_000, 16_000, 8_000]` — was `[128_000, ...]`. `DEFAULT_FALLBACK_CONTEXT` follows tier[0], so unknown models now default to 256K. The stale `128000` literal in the OpenRouter metadata-miss path is replaced with `DEFAULT_FALLBACK_CONTEXT` for consistency.

## Repro (from #15779)

```yaml
custom_providers:
  - name: my-custom-endpoint
    base_url: https://example.invalid/v1
    model: gpt-5.5
    models:
      gpt-5.5:
        context_length: 1050000
```

`/model gpt-5.5 --provider custom:my-custom-endpoint` → previously "Context: 128,000", now "Context: 1,050,000".

## Tests

- `tests/hermes_cli/test_custom_provider_context_length.py` — new file, 19 tests covering the helper, step-0b integration, and the 256K tier invariants
- `tests/hermes_cli/test_model_switch_context_display.py` — added regression tests for #15779 through the display resolver
- `tests/gateway/test_session_info.py` — updated default-fallback assertion (128K → 256K)
- `tests/agent/test_model_metadata.py` — updated tier assertions for the new top tier

tests/hermes_cli/test_model_switch_context_display.py

@@ -88,3 +88,61 @@ class TestResolveDisplayContextLength:
                 model_info=fake_mi,
             )
         assert ctx == 128_000
+
+    def test_custom_providers_override_honored(self):
+        """Regression for #15779: /model switch onto a custom provider must
+        surface the configured per-model context_length, not the 128K/256K
+        fallback.
+        """
+        custom_provs = [
+            {
+                "name": "my-custom-endpoint",
+                "base_url": "https://example.invalid/v1",
+                "models": {"gpt-5.5": {"context_length": 1_050_000}},
+            }
+        ]
+        # Real resolver call — no mock — so the override path is exercised
+        # through agent.model_metadata.get_model_context_length.
+        from unittest.mock import patch as _p
+        from agent import model_metadata as _mm
+        with _p.object(_mm, "get_cached_context_length", return_value=None), \
+             _p.object(_mm, "fetch_endpoint_model_metadata", return_value={}), \
+             _p.object(_mm, "fetch_model_metadata", return_value={}), \
+             _p.object(_mm, "is_local_endpoint", return_value=False), \
+             _p.object(_mm, "_is_known_provider_base_url", return_value=False):
+            ctx = resolve_display_context_length(
+                "gpt-5.5",
+                "custom",
+                base_url="https://example.invalid/v1",
+                api_key="k",
+                custom_providers=custom_provs,
+            )
+        assert ctx == 1_050_000, (
+            "custom_providers[].models.gpt-5.5.context_length=1.05M must win "
+            "over probe-down fallback"
+        )
+
+    def test_custom_providers_trailing_slash_insensitive(self):
+        """Base URL comparison must tolerate trailing-slash differences
+        between config.yaml and the runtime value.
+        """
+        custom_provs = [
+            {
+                "base_url": "https://example.invalid/v1/",
+                "models": {"m": {"context_length": 400_000}},
+            }
+        ]
+        from unittest.mock import patch as _p
+        from agent import model_metadata as _mm
+        with _p.object(_mm, "get_cached_context_length", return_value=None), \
+             _p.object(_mm, "fetch_endpoint_model_metadata", return_value={}), \
+             _p.object(_mm, "fetch_model_metadata", return_value={}), \
+             _p.object(_mm, "is_local_endpoint", return_value=False), \
+             _p.object(_mm, "_is_known_provider_base_url", return_value=False):
+            ctx = resolve_display_context_length(
+                "m",
+                "custom",
+                base_url="https://example.invalid/v1",  # no trailing slash
+                custom_providers=custom_provs,
+            )
+        assert ctx == 400_000