use -puny in denominator of scaling ratio for flux limiting

[eclare108213]

• [x] Short (1 sentence) summary of your PR: Fix the sign of puny in the denominator of the ratio used to scale heat fluxes
• [x] Developer(s): @njeffery
• [x] Suggest PR reviewers from list in the column to the right.
• [ ] Please copy the PR test results link or provide a summary of testing completed below. Bug fix tested in E3SM by @njeffery.
• How much do the PR code changes differ from the unmodified code?
  • [ ] bit for bit
  • [x] different at roundoff level
  • [ ] more substantial
• Does this PR create or have dependencies on CICE or any other models?
  • [ ] Yes
  • [x] No
• Does this PR add any new test cases?
  • [ ] Yes
  • [x] No
• Is the documentation being updated? ("Documentation" includes information on the wiki or in the .rst files from doc/source/, which are used to create the online technical docs at https://readthedocs.org/projects/cice-consortium-cice/.)
  • [ ] Yes
  • [x] No, does the documentation need to be updated at a later time?
    • [ ] Yes
    • [x] No
• [x] Please provide any additional information or relevant details below: Bottom and lateral heat fluxes are limited by the total amount of heat available for melting, and all of these terms are negative. In particular, fbot+fside<0. Therefore the value of puny needed to keep the denominator from being too small should also be negative.

[apcraig]

Just want to add a note that this changes answers in all cases.