[TRACKING ISSUE] Set cluster connect timeout to infinite by default

The tracking issue for the Java side PR.

Set cluster connect timeout to infinite by default

See hazelcast/hazelcast#18094 for details.

---

After an internal discussion, we have decided to set the cluster
connect timeout to infinite by default. Also, the multiplier
for the sleep time between consecutive connect attempts is changed
to 1.05 from 1.0. The new sleep times between connection
attempts can be seen in the following graph. https://www.desmos.com/calculator/hl7rnbk4gf
(x-axis: attempt number, y-axis: sleep time in seconds)

Also, updated some logging statements.

Note that, this is a breaking change for the users who rely on the client to stop
trying to connect cluster after some time, but we decided