Is there floating point type with more than 64 bits?
IEEE 754, the primary standard for floating point, defines larger types, including 128-bit and 256-bit, but at this time Rust does not support those larger types.
Is support planned?
Here's some recent discussion about it.
What is reason?
For the reason, I don't think there's a hard reason to not add them to the language. We already have i128, so why not f128? You can write a RFC, and find a best way to include them into the language by discussions and bikesheddings.
This is the RFC for the i128. I hope it helps you to write one.