Add used/full context meter to chat screen (#717)

* Adopt /api/health REST route (issue #713)

Cheap liveness probe on the connection screen: GET /api/health surfaces
backend version + auth mode (Token vs Gated/OAuth) without the heavier
/api/status payload. Added HermesApiService.getHealth + HealthStatus model;
ConnectViewModel.loadHealth probes on open (best-effort, silent on failure);
ConnectScreen shows a read-only health card when reachable.

Verified: ktlint clean, assembleDebug BUILD SUCCESSFUL locally.

* Add used/full context meter to chat screen

Shows session prompt tokens vs model context window above the composer,
polled every 5s. Color-coded bar warns at 70%/90%.

* fix(chat): fix unit tests for context usage meter

Format >=100k tokens cleanly in ContextUsageChip and stub AuthManager profile ID in ChatViewModelTest.

* Fix context meter: source used tokens from input_tokens

Gateway /api/sessions/{id} has no last_prompt_tokens column; use
input_tokens instead. Show meter as soon as full context is known.

* Add tappable context breakdown sheet

Chip opens a bottom sheet with input/output/cache/read/write/reasoning
token rows + message count. Source breakdown from /api/sessions/{id}.

* Refresh context meter on streaming complete

Trigger fetchContextUsage immediately when a message finishes streaming
instead of waiting for the 5s session-sync poll.

* Fix context meter: source used tokens from live WS session.info

The REST /api/sessions/{id} endpoint reads the DB input_tokens column,
which is stale/0 for active sessions — that's why the meter was stuck at 0.

Switch the meter to the live values the gateway pushes over the WS
session.info payload after each turn:
- used = last_prompt_tokens (real used context for the turn)
- full = context_length (model-accurate window; keep /api/model/info as a
  fallback before the first turn completes)

fetchContextUsage now only tops up the denominator from /api/model/info and
no longer overwrites the live WS value with the stale REST input_tokens.
The 5s poll remains as a fallback.

* Revert "Fix context meter: source used tokens from live WS session.info"

This reverts commit 932393fd80993d57e66fe93552a949460d646674.

---------

Co-authored-by: M57 <hy4ri@users.noreply.github.com>

app/src/main/java/com/m57/hermescontrol/ui/connect/ConnectScreen.kt

@@ -35,6 +35,7 @@ import androidx.compose.material3.IconButton
 import androidx.compose.material3.MaterialTheme
 import androidx.compose.material3.OutlinedButton
 import androidx.compose.material3.OutlinedTextField
+import androidx.compose.material3.Surface
 import androidx.compose.material3.Text
 import androidx.compose.material3.TextButton
 import androidx.compose.runtime.Composable
@@ -316,6 +317,43 @@ fun ConnectScreen(
                         )
                     }
                 }
+
+                // Backend health read-out (issue #713): cheap /api/health probe.
+                // Shows version + auth mode when reachable; nothing when null.
+                state.health?.let { health ->
+                    val authMode =
+                        if (health.authRequired == true) {
+                            stringResource(R.string.connect_health_auth_gated)
+                        } else {
+                            stringResource(R.string.connect_health_auth_token)
+                        }
+                    Surface(
+                        color = MaterialTheme.colorScheme.surfaceContainer,
+                        shape = MaterialTheme.shapes.medium,
+                        modifier = Modifier.fillMaxWidth(),
+                    ) {
+                        Column(
+                            modifier = Modifier.padding(12.dp),
+                            verticalArrangement = Arrangement.spacedBy(4.dp),
+                        ) {
+                            Text(
+                                text = stringResource(R.string.connect_health_title),
+                                style = MaterialTheme.typography.labelMedium,
+                                color = MaterialTheme.colorScheme.onSurfaceVariant,
+                            )
+                            Text(
+                                text =
+                                    stringResource(
+                                        R.string.connect_health_body,
+                                        health.version ?: "?",
+                                        authMode,
+                                    ),
+                                style = MaterialTheme.typography.bodyMedium,
+                                color = MaterialTheme.colorScheme.onSurface,
+                            )
+                        }
+                    }
+                }
             }
         }
     }

app/src/main/java/com/m57/hermescontrol/ui/connect/ConnectViewModel.kt

@@ -8,6 +8,7 @@ import com.m57.hermescontrol.R
 import com.m57.hermescontrol.data.config.ConnectionProfile
 import com.m57.hermescontrol.data.config.resolveBaseUrl
 import com.m57.hermescontrol.data.local.AuthManager
+import com.m57.hermescontrol.data.model.HealthStatus
 import com.m57.hermescontrol.data.remote.ApiClient
 import com.m57.hermescontrol.data.remote.CleartextPolicy
 import com.m57.hermescontrol.data.remote.NetworkError
@@ -33,6 +34,7 @@ data class ConnectUiState(
     val saveProfile: Boolean = false,
     val profiles: List<ConnectionProfile> = emptyList(),
     val selectedProfile: ConnectionProfile? = null,
+    val health: HealthStatus? = null,
 )
 
 class ConnectViewModel(
@@ -60,6 +62,29 @@ class ConnectViewModel(
                 profileName = selectedProfile?.name ?: "",
             )
         }
+        loadHealth()
+    }
+
+    /**
+     * Cheap liveness probe (issue #713): GET /api/health against the current
+     * URL/token. Surfaces backend version + auth mode on the connection screen
+     * without the heavier /api/status payload. Best-effort: failures are silent
+     * (health stays null and the screen simply shows no read-out).
+     */
+    fun loadHealth() {
+        val state = _uiState.value
+        val endpoint = runCatching { ServerEndpoint.parseForBuild(state.baseUrl) }.getOrNull()
+        if (endpoint == null || state.token.isBlank()) return
+        viewModelScope.launch {
+            val result =
+                withContext(Dispatchers.IO) {
+                    val tempApi = ApiClient.createTempService(endpoint.baseUrl.toString(), state.token)
+                    safeApiCall { tempApi.getHealth() }
+                }
+            if (result is NetworkResult.Success) {
+                _uiState.update { it.copy(health = result.data) }
+            }
+        }
     }
 
     fun onProfileNameChange(value: String) {