fix: pin postgres image version; fix default db host

The former fixes some issues when postgres 18 is used AFAICT; 17 is what is used in prod: https://github.com/mozilla-services/cloudops-infra/blob/master/projects/lando/tf/modules/postgres/variables.tf#L10-L12

Errors I was seeing were:
```
~/repos/lando main ?1 ❯ make test                                                                                                                                                            х INT
/usr/bin/docker compose run --rm lando lando tests
WARN[0000] The "UID" variable is not set. Defaulting to a blank string.
WARN[0000] The "GID" variable is not set. Defaulting to a blank string.
[+] Creating 1/1
 ✔ Container lando-redis-1  Running                                                                                                                                                           0.0s
[+] Running 1/1
 ✔ Container lando-db-1  Started                                                                                                                                                              0.2s
dependency failed to start: container lando-db-1 exited (1)
make: *** [Makefile:35: test] Error 1
~/repos/lando main ?1 ❯ docker logs lando-db-1
Error: in 18+, these Docker images are configured to store database data in a
       format which is compatible with "pg_ctlcluster" (specifically, using
       major-version-specific directory names).  This better reflects how
       PostgreSQL itself works, and how upgrades are to be performed.

       See also docker-library/postgres#1259

       Counter to that, there appears to be PostgreSQL data in:
         /var/lib/postgresql/data (unused mount/volume)

       This is usually the result of upgrading the Docker image without
       upgrading the underlying database using "pg_upgrade" (which requires both
       versions).

       The suggested container configuration for 18+ is to place a single mount
       at /var/lib/postgresql which will then place PostgreSQL data in a
       subdirectory, allowing usage of "pg_upgrade --link" without mount point
       boundary issues.

       See docker-library/postgres#37 for a (long)
       discussion around this process, and suggestions for how to do so.
```

The latter fixes a bug that I presume nobody else is hitting because they've overriden the db hostname in `.env`.

Author: bhearsum

compose.yaml

@@ -1,6 +1,6 @@
 services:
   db:
-    image: postgres
+    image: postgres:17
     environment:
       POSTGRES_USER: postgres
       POSTGRES_PASSWORD: postgres

src/lando/settings.py

@@ -139,7 +139,7 @@
         "NAME": os.getenv("DEFAULT_DB_NAME", "postgres"),
         "USER": os.getenv("DEFAULT_DB_USER", "postgres"),
         "PASSWORD": os.getenv("DEFAULT_DB_PASSWORD", "postgres"),
-        "HOST": os.getenv("DEFAULT_DB_HOST", "lando.db"),
+        "HOST": os.getenv("DEFAULT_DB_HOST", "db"),
         "PORT": os.getenv("DEFAULT_DB_PORT", "5432"),
     }
 }