feat: congestion window control

[AnoldGH]

No description provided.

[AnoldGH]

To inform congestion control about packet losses we need to modify/append ExpressR API. One approach is to introduce an additional retry callback function triggered upon timeouts instead of immediately retrying.

[AnoldGH]

Issues Identified:

1. Decoupled Retransmission and Congestion Control:
   The current design handles retransmission via ExpressR, which is decoupled from congestion window size. This causes congestion mitigation to happen slowly and introduces overheads in strategies (e.g. to ensure cwnd increases once per RTT).

2. Limited Reliability Guarantee
   ExpressR retries a fixed number of times before discarding the result. This offers only partial reliability and likely leads to the failure observed in this experiment.

3. Interest Generation tied to seqnum:
   New interests are expressed strictly based on sequence number ordering, which does not align well with a QUIC-like window model.

I think a refactoring is needed to address these limitations. Hence I will close this PR and submit a new one when more work is ready.

[pulsejet]

Sorry I haven't had time yet to do a proper review yet, but this looks on the right track!

Please feel free to propose changes to existing APIs like expressr if needed