What's the difference here?
Are stateful IPv6 entries retained in the DHCPv6 server?
Stateful autoconfiguration of IPv6 is the equivalent to the use of DHCP in IPv4. It requires a DHCPv6 service to provide the IPv6 address to the client device and that both client device and server maintain the "state" of that address (i.e. lease time, etc).
Stateless autoconfiguration of IPv6 allows the client device to self-configure its IPv6 address and routing based on the router advertisements.
A network can use both stateful and stateless autoconfiguration at the same time, they are not mutually exclusive.
If you are interested in more detailed information, I would highly recommend you read RFC4862.
With stateless autoconfiguration the clients get prefixes and routes from router advertisements and choose their own addresses within the prefixes, these address may be based on the MAC address and/or selected randomly. Many clients will implement "privacy extensions" where they cycle through randomly assigned addresses frequently. Additional configuration may be provided either through extensions in the router advertisements or through stateless DHCPv6.
With stateful DHCPv6 addresses are allocated in much the same way as with IPv4 DHCP. It is up to the DHCP server to hand out addresses (or not) as it sees fit and to keep track of which client has which address.
The big downside of stateless, especially with privacy extensions is it makes it much more effort to track and/or block abuse. With stateful DHCP (v4 or v6) a client is likely to keep a stable IP as long as it stays on the same network, with stateless and privacy extensions it will jump around all over the place.
The big downside of stateful DHCPv6 is that andriod doesn't support it. So on a wireless segment that will be used by phones/tablets you are basically forced to allow stateless autoconfiguration if you want all clients to have v6 access.
It is technically possible to use both on the same network at the same time, I would question whether it is sensible though.