Verbosity improvements

[RobinD42]

Switch most mpb_verbosity checks to be >=1, and change the linmin messages to >=2

See NanoComp/meep#1362 and NanoComp/meep#1388

[thchr]

This seems to have the same problem that I brought up in #119 (comment) and fixed in #123?

Specifically, if the default verbosity level for MPB is 2, then this prints the linmin messages by default - which is a lot of extra stuff. That seems to be the current situation, cf. Travis reports for this PR.

I'm guessing this works well with Meep, if it's reducing the verbosity level by 1 when calling out to MPB; but it's not too nice for working with MPB on its own.

[thchr]

@RobinD42: another consequence of this issue is that e.g. the Meep logs now exceed the allowable length on Travis due to repeated prints of "linmin: converged after ...", see e.g. https://travis-ci.org/github/NanoComp/meep/jobs/741695728#L79272

Wouldn't it better to roll back these changes to linmin printing?

[stevengj]

Wouldn't it better to roll back these changes to linmin printing?

Yes, we can do that; it may have been an inadvertent change. Can you make a PR?

[thchr]

Sure (I probably won't get around to it until the second half of next week though)!

[stevengj]

@thchr, we changed the MPB default verbosity to 1 in #130, which should fix this. (It was already 1 in the Python interface.)