Add README

Bump experimental `/img` to Dal E 3

.gitignore

 work/
 .env
 *.swp
+*.backup

README.md

+#### Running
+
+* `./debug-server.sh`
+* Browse to [http://localhost:5000]
+
+#### Prompt files
+
+Prompt template is at [templates/prompt.txt](templates/prompt.txt).
+
+The sources of the fields are:
+
+* **{{user}}** — One of the files at [static/users](static/users), describing current user.
+* **{{schema}}** — [static/chat.schema.json](static/chat.schema.json) (after dereference). Describing the syntax [and some semantics] of JSON replies we expect from the assistant.
+* **{{example}}** — [static/chat-example.json](static/chat-example.json)
+* **{{img_enum}}** — Available images (including text description). See [here](http://localhost:5000/enum/img) (while server is running)

app.py

     else:
         prompt = request.form["prompt"]
         alt = prompt
-        response = openai.Image.create(prompt=prompt, n=1, size="1024x1024")
+        response = openai.Image.create(prompt=prompt, model="dall-e-3", n=1, size="1024x1024")
         src = response['data'][0]['url']
     return render_template("image.html", src=src, alt=alt, prompt=prompt)