clippy::cast-possible-truncation generates warnings

[LPardue]

We don't normally run the rule -W clippy::cast-possible-truncation but if I try we see a few messages such as :

warning: casting u64 to usize may truncate the value on targets with 32-bit wide pointers
   --> src/lib.rs:950:13
    |
950 |             config.local_transport_params.initial_max_streams_bidi as usize,
    |             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    |
    = note: `-W clippy::cast-possible-truncation` implied by `-W clippy::pedantic`
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#cast_possible_truncation

warning: casting u64 to usize may truncate the value on targets with 32-bit wide pointers
   --> src/lib.rs:954:13
    |
954 |             config.local_transport_params.initial_max_streams_uni as usize,
    |             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#cast_possible_truncation

warning: casting u64 to usize may truncate the value on targets with 32-bit wide pointers
    --> src/lib.rs:1173:13
     |
1173 |             b.get_varint()? as usize
     |             ^^^^^^^^^^^^^^^^^^^^^^^^

[ghedo]

There's probably a lot of those messages, and in most cases they don't indicate bugs (which is why the lint is not enabled by default).

Though the first two examples you mentioned are probably bugs, so I made #122 to fix them.

[LPardue]

Cool yeah clippy pedantic is pretty spammy and I dont think we should respect all of its moaning. I suspect some of them are from the recent 32bit changes but will assume that these changes fix possible problems.