Hardcoded precision with overwrite in make_ops

[stloufra]

Once generated (by default) there is hardcoded:

─────┬───────────────
     │ File: Source/make_opts
─────┼───────────────
   1 │ override FPTYPE=m
   2 │ pdlabel1=
...

which later prevents compiling the check_sa.cc with e.g. make ... FPTYPE=f to be taken into account. However, there is already default option implementation:

─────┬───────────────
     │ File: src/cudacpp_config.mk
─────┼───────────────
...
  18 │ ifeq ($(FPTYPE),)
  19 │   # OLD DEFAULT UP TO v1.00.00 INCLUDED (inconsistent with default
     │  floating_type='m' in run_card.dat)
  20 │   ###override FPTYPE = d
  21 │   # NEW DEFAULT (#995) AS OF v1.00.01 (now consistent with default
     │  floating_type='m' in run_card.dat)
  22 │   override FPTYPE = m
  23 │ endif
...

I traced the source back to:

──────────────────────
     │ File: ./PLUGIN/CUDACPP_SA_OUTPUT/launch_plugin.py
─────┼────────────────
  75 │     def reset_makeopts(self, old_value, new_value, name):
  76 │         if not hasattr(self, 'path'):
  77 │             raise Exception
  78 │         if name == 'floating_type':
  79 │             common_run_interface.CommonRunCmd.update_make_opts_full({'override FPTYPE': new_value})
  80 │         else:
  81 │             raise Exception
  82 │         Sourcedir = pjoin(os.path.dirname(os.path.dirname(self.path)), 'Source')
  83 │         subprocess.call(['make', 'cleanall'], cwd=Sourcedir, stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)
  84 │
  85 │     def default_setup(self):
  86 │         super().default_setup()
  87 │         self.add_param('floating_type', 'm', include=False, hidden=True,
  88 │                        fct_mod=(self.reset_makeopts,(),{}),
  89 │                        allowed=['m','d','f'],
  90 │                        comment='floating point precision: f (single), d (double), m (mixed: double for amplitudes, single for colors)'
  91 │                        )

This prevents development manually within the check_sa.cc and using make calls.

[roiser]

At a first glance I would propose to remove the override otherwise make would never be able to take precedence over the Makefile option.