[Statistics] Improve SRT core statistics documentation

[mbakholdina]

Improve SRT core statistics documentation.

Related Issues: #1215, #1213.
Related PRs: #1199, #1216.

First Priority

• Summary table: Fill in the remain empty cells of the receiver.

• Available for sender / Available for receiver: Check that detailed description of all the statistics contains either "Available for sender" or "Available for receiver"

• Fix the links.

• msRTT, mbpsBandwidth:

• Check how the statistics is calculated on both sides and add corresponding comments plus add a comment regarding estimations exchange between rcv and snd through ACK packets,
• Statistics are calculated independently on each side (receiver and sender) and are not exchanged between peers unless explicitly stated. - Update this sentence from the very beginning to reflect the idea of "unless explicitly stated".

• Not all the statistics have mentioning that it's data packets, e.g., pktRecvTotal - correct this.

• Formulas for calculation of loss ratio, dropped packets ratio, the number of recovered/unrecovered packets, etc.

• Improve the description of pktRcvLossTotal statistic.

• Improve the description of pktSndDropTotal, pktRcvDropTotal statistics.

• Instantaneous statistics: Check how moving average is calculated for all the instantaneous statistics. See issue [BUG] Statistics always return instantaneous value for send and recv buffers #1272.

Second Priority

• pktRcvLossTotal: See issue [Statistics] Correct the way of pktRcvLossTotal statistic calculation #1212.

• pktRetransTotal, byteRetransTotal: See issue [Statistics] Change the name of statistic "pktRetransTotal" to "pktSndRetransTotal" #1209. Check corresponding interval-based stats.

• pktRcvRetransTotal, byteRcvRetransTotal (both not implemented): See issue [Statistics] Add the calculation of pktRcvRetransTotal and byteRcvRetransTotal statistics #1208. Check corresponding interval-based stats.

• pktRcvBelatedTotal (not implemented): See issue [Statistics] Implement pktRcvBelatedTotal statistic #1214.

• pktSndDropTotal:

• Revise the following text: "The total delay before TLPKTDROP mechanism is triggered consists of the SRTO_PEERLATENCY, plus SRTO_SNDDROPDELAY, plus 2 * the ACK interval (default ACK interval is 10 ms). The delay used is the timespan between the very first packet and the latest packet in the sender's buffer."

• Better to update the RFC with this information and from statistics documentation make a link to the RFC.

• FEC related statistics:

• Would be nice to revise how FEC statistics is calculated and whether it affects the regular SRT statistics.
• Make a note (if does not exist) in FEC documentation: Packet filter control packets are SRT DATA packets. Why not make FEC packets SRT CONTROL?
• Issue [BUG] Bandwidth Estimation with SRT 1.4.1 is always 0 #1021

[mbakholdina]

@ethouris Could you please help me with improving the current description for pktSndDropTotal, pktRcvDropTotal statistics of individual socket. I suggest to create a PR from the current master and correct only these 2 statistics. Please make me a reviewer.