1.  IMO/E “ejection pattern” is essentially meaningless = don’t make adjustments based on that alone.  If the brass comes out of he gun, your ejection pattern is good.

2.  How does the gun run?  Reliable feed/extraction = don’t mess with it.

3.  The fractional reduction in gas yo might make with an adjustable gas block for example won’t materially change he amount of “felt” gas.

4.  Search for the charging handle “RTV mod” or get something like a PRI Gasbuster charging handle.

A similar/recent thread.

There are the two sides to the story:

1.  Don’t worry about it.

2.  Try to fix it.

If with full charging handle pull, face of the bolt is stopping at the correct distance, hence 1/8"~1/4" in front of the ejection port, and extractor is sharp enough, the extractor spring strong enough to keep the bolt pinned to the bolt face all the way back (not pulled off the spent case rim during the spent case pull out of the chamber), then angle change in ejection path out of the ejection port (before the spent case hits the upper receiver deflector), is all about the amount of dead blow stall time that buffer has at the back of stroke.

So heaver buffer than needed, just slows down the unlock timing, reduced the amount of speed that the buffer has to strike the back of receiver extension to decrease the rear ward blow to the back of the tube, which can cause forward ejection.

Some token with a buffer that is too light, which causes the bolt to unlock too fast, but since it takes more forced to pull the spent cases out of the chamber since the bore residual pressure is too high, this slows the bolt rearward travel as well.  More so, the faster the bolt unlock prematurely, the strong that the extractor spring needs to be, so the extractor is not pulled off the spent case rim on the pull.

As for suppressor, it creates more back pressure in the bore, so not only do this increase the gas pulse amount to open the bolt sooner, but due to the slightly longer duration of back blast in the bore, hold the residual pressure slightly slower in bleed off, which puts more pressure on the spent case to pressure weld it in the chamber as well.

So on say the M4, you not only have heaver buffer to slow the bolt unlock back down, but M4 extra tension extractor spring so the extractor holds onto the spent case rim  on the bolt (holds in on the bolt all the way back until dead blow effect of the buffer.

Then we get to the elephant in the room, being that the 10.5 barrel is going to be very ammo sensitive to start with.
Bluntly put, even with a moderator (to make the rifle sound like a 20" barrel rifle) on the first XM-177's with 10.5" barrels, rig was just too ammo sensitive for even the select few military ammo's of the time. The quick solve to the problem, was the Xm-177E2, which got a 11.5" barrel isntead.

So since you running from weaker 223 ammo (55K working pressure), all the way up to full Nato loads of 62K working pressure in the 10.5 rig, and it is ejecting the spent cases out to begin with (without jams), would stop worrying about the ejection angles out to the ejection port to begin with.

On the other hand, if you want to settle into Just One ammo type to use in the rig alone, then we can tune the rig to put the spent case all in the same pile out the rifle at the 4:30 position.

No, Tula ammo is steel case with Poly or lacquer coating on it to keep the cases from rusting, so you have a problem with either the chamber not being chamber brush scrubbed cleaned with CLP by hand, then dried with dry patches, before the rifle is first shot, and then scrubbed again every cleaning.

Or, chamber surface not mirror smooth finished, but reamer chattered to hell instead.
Hence if you can run brass case ammo, but not the steel case, then chamber could just be on the tighter side, slightly rougher side, and may just take some break in with brass case ammo until the chamber walls do self polish out enough from life fire to run the steel case ammo.

And yes, chamber has to be chamber brush scrubbed with CLP by hand, since its the chamber brush that is removing the poly or lacquer powderized fouling coming from the Steel case ammo, since there is not a standard cleaning agent that you can use on the rifle that will disolve that powderized fouling, and not melt the plastic butt stock/forearm/pistol grip right off the rifle at the same time.

As for the CLP your using for the chamber scrubbing, it just suspending the fouling as you scrub out the debris off the chamber walls, and will be removed when you run the clean patches and Q tips to clean the barrel extension void areas.

As for steel case ammo once the rifle has been cleaned and lubed on the wetter side correctly with CLP, if the rifle does get to the point that the case are stuck to the chamber walls, spent case left in the chamber as the extractor slips off the rim, then it time to clean the chamber again.  Running the upper bearing surfaces a tad wetter with CLP will help with the steel case ammo.

Also to note, if you are using Hoppe's cleaning solvent on the rifle, stop using it.  It leaves a protective coating behind, and this coating will collect the poly/laquer powderized fouling that is being scrapped off the case even faster.  So bore itself gets cleaned with something like Sweets that will not leave a protective coating behind, then CLP and chamber brush used to clean the chamber section of the barrel, with the fouled CLP pushed from chamber to muzzle out of the rifle.  Even when you think that you have the chamber and bore dry, there is enough rements of CLP still on the surfaces for short term storage of the rig to prevent rusting.   The rest of the rifle, cleaned first CLP, that Fouled CLP wiped away, then upper receiver bearing areas get a light coat of CLP to lube the the rifle.

So simply, it a clean and lubing problem to start with, or could be a rougher than normal chamber isntead.  99% of the time if you can not see reamer grooving in the chamber once cleaned, it the cleaning and lubing you are doing wrong instead (technique or wrong solvents and lube).
Note, CLP has a cleaning agent in it.