flux resource list output confusing with multiple queues

[ryanday36]

The queue column of flux resource list can be confusing and hard to read when there are multiple overlapping queues. e.g.

[day36@rzadams1001:~]$ flux resource list
     STATE QUEUE      NNODES   NCORES    NGPUS NODELIST
      free plarge,pde      2      192        8 rzadams[1096,1122]
 allocated pdebug,pal     12     1152       48 rzadams[1007-1009,1011-1012,1015-1016,1020-1021,1023,1025-1026]
 allocated plarge,pde     77     7392      308 rzadams[1027-1032,1049-1063,1065-1095,1097-1121]
      down pdebug,pal     12     1152       48 rzadams[1003-1006,1010,1013-1014,1017-1019,1022,1024]
      down plarge,pde     23     2208       92 rzadams[1033-1048,1064,1123-1128]
[day36@rzadams1001:~]$

This is, admittedly, made worse by our decision to give two of the queues the same initial three letters (pdev and pdebug), but it's generally difficult to determine whether there are nodes available in a specific queue. A separate line for each queue might help make this more clear, although there might be other possible solutions as well.

[grondo]

Open to any suggestions. Unfortunately, flux resource list currently organizes output by common sets of resources, it is not, for example, iterating through configured queues and printing resources that way. I could look into doing that though of course resources would be duplicated between rows (which may cause confusion if a user sees 2 nodes free in pdebug and also 2 nodes free in plarge)

I wonder if an immediate improvement would be to extend the width of the QUEUE column on systems where there are overlapping queues. This can be done via a FLUX_RESOURCE_LIST_FORMAT_DEFAULT environment variable. Unfortunately we don't allow the default format to be overridden by config (though that could be changed, I think there were arguments against it)