Possible typo in optimal value of `qap10` in README.md

[napinoco]

In the README.md, the optimal value of qap10 is given as -1.093e+01.
However, based on the experimental results reported in several papers, such as the SDPT3 User Guide, and my own verification, I believe the correct optimal value should be -1.093e+03.

[siko1056]

Thank you @napinoco for pointing out this point. Actually it is the original value reported by Borchers in 1999: https://doi.org/10.1080/10556789908805769 (page 688)

However, we share your observation and maybe I can include a footnote to this value:

• https://vsdp.github.io/benchmark/2018_12/SDPLIB_SYS2.html (my past research results, search for "qap10", used Mosek solver)
• https://www.tuhh.de/ti3/jansson/vsdp2012/VSDP2012Guide.pdf (my colleagues found the same in 2012, with SDPT3 solver)
• https://blog.nus.edu.sg/mattohkc/softwares/sdpt3/ contains the guide https://bpb-us-w2.wpmucdn.com/blog.nus.edu.sg/dist/b/11036/files/2019/10/guide4-0-draft.pdf (stating your value as well)