slow-odgi: tidying

[anshumanmohan]

In addition to #33, some other high-level things that should be done after merging https://github.com/cucapra/pollen/pull/29 and https://github.com/cucapra/pollen/pull/31:

• [x] Since beginning to write slow-odgi, I have discovered how to pipe odgi commands. Do that more; cut down on the number of temp files. Maybe this will let me avoid the temp storm (temp-est 🥁 🥁 ✨) I sometimes have, where temp files created when testing one command then become targets for the next command.
• [x] Test against more of the graphs that are already in the odgi repo. I presume the present selection that we fetch with make fetch have been chosen because they are of a reasonable size? @susan-garry?
• [x] The graphs that we fetch frequently fail to actually test odgi commands rigorously. I have in the back of my mind a bit of a zoo of small graphs that could be added, either to the odgi repo or elsewhere, that, when tested against, would run the commands through their paces.
• [x] Refactor! There is some known duplication across branches, and sometimes I have done the same thing in different ways in different commands.

[sampsyo]

Test against more of the graphs that are already in the odgi repo. I presume the present selection that we fetch with make fetch have been chosen because they are of a reasonable size?

FWIW, the files checked into the odgi repo are all pretty small. As of this writing, the ones we fetch are: https://github.com/cucapra/pollen/blob/af62febce2a4ae1afe427ecf0ce89cde2c55e416/Makefile#L1

Measuring those sizes, we have:

$ for name in t k note5 overlap q.chop DRB1-3123 LPA ; do ls -lh $name.gfa ; done
-rw-r--r--@ 1 asampson  staff   491B Nov 24  2021 t.gfa
-rw-r--r--@ 1 asampson  staff   561B Nov 24  2021 k.gfa
-rw-r--r--@ 1 asampson  staff   168B Nov 24  2021 note5.gfa
-rw-r--r--@ 1 asampson  staff   362B Nov 24  2021 overlap.gfa
-rw-r--r--@ 1 asampson  staff   1.4K Nov 24  2021 q.chop.gfa
-rw-r--r--@ 1 asampson  staff   452K Nov 24  2021 DRB1-3123.gfa
-rw-r--r--@ 1 asampson  staff   1.5M Nov 24  2021 LPA.gfa

So, they range from trivial to tiny. The exception, of course, is chr8, which is properly big (and not in the repo for that reason):

$ ls -lh chr8.pan.gfa
-rw-r--r--  1 asampson  staff   3.9G Feb  9 07:48 chr8.pan.gfa