dev stack: pip-install local omni-pca on HA startup

Replace the brittle bind-mount-over-site-packages trick with a proper
``pip install --no-deps /opt/omni-pca-src`` in the HA container's
entrypoint. This gives HA a real ``omni_pca-2026.5.10.dist-info`` so
the manifest's requirement check passes, plus the v1 subpackage that's
not in the published wheel yet (omni-pca==2026.5.10 isn't on PyPI).

Before: ``--force-recreate`` broke the dev stack because the bind mount
overlaid the package contents but left no dist-info, and HA's uv-based
installer can't fetch omni-pca from PyPI.

After: container recreate just works. ``docker compose restart
homeassistant`` re-installs from the latest local source on every
start, so HA + library are always in sync with the working tree.

Header comments updated to mention the real-panel (UDP/v1) config-flow
fields alongside the existing mock-panel ones.

dev/docker-compose.yml

@@ -6,15 +6,27 @@
 #   make dev-logs    # tail HA logs
 #   make dev-down    # stop and clean
 #
+# On every container start the HA service pip-installs the local
+# `omni-pca` library from ../ into site-packages (the version pinned in
+# the integration manifest isn't on PyPI yet, and we want our latest
+# v1/ subpackage available either way). Source changes in src/omni_pca
+# require a ``docker compose restart homeassistant`` to take effect.
+#
 # Once running, open http://localhost:8123 and:
 #   1. Onboard with any name / location.
 #   2. Settings -> Devices & Services -> Add Integration ->
 #      "HAI/Leviton Omni Panel".
-#   3. Use:
-#        host             host.docker.internal
-#        port             14369
-#        controller_key   000102030405060708090a0b0c0d0e0f
-#      (matches scripts/run_mock_panel.py defaults)
+#   3. Use one of:
+#        Mock panel (TCP):
+#          host             host.docker.internal
+#          port             14369
+#          transport        TCP
+#          controller_key   000102030405060708090a0b0c0d0e0f
+#        Real panel (UDP, v1 wire protocol):
+#          host             <panel IP, e.g. 192.168.1.9>
+#          port             4369
+#          transport        UDP
+#          controller_key   <32 hex chars from the panel's .pca file>
 
 services:
   mock-panel:
@@ -40,9 +52,25 @@ services:
     volumes:
       - ./ha-config:/config
       - ../custom_components/omni_pca:/config/custom_components/omni_pca:ro
+      # Make the whole library project (pyproject + src/ + dist/) available
+      # so the entrypoint override below can pip-install from local source
+      # before /init starts. This gives HA real dist-info for
+      # ``omni-pca==2026.5.10`` (which isn't on PyPI yet) and ensures the
+      # v1 subpackage is present.
+      - ../:/opt/omni-pca-src:ro
     ports:
       - "8123:8123"
     extra_hosts:
       - "host.docker.internal:host-gateway"
     environment:
       - TZ=America/Boise
+    # HA's image entrypoint is /init (s6-overlay). We pre-install our
+    # local library against site-packages so HA's manifest-requirement
+    # check finds it, then exec /init normally.
+    entrypoint:
+      - sh
+      - -c
+      - |
+        set -e
+        pip install --quiet --no-deps --upgrade /opt/omni-pca-src
+        exec /init